Publications
Income and Expenditure Accounts Technical Series
Canadian Tourism Satellite Account, 2004
Appendix B Sources and methods1
Archived Content
Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.
Data sources
Several main data sources are used for the Canadian Tourism Satellite Account (CTSA). Demand estimates are derived from two main sources, the Canadian Travel Survey (CTS) and the International Travel Survey (ITS), both conducted by Statistics Canada. The CTS provides data for domestic business and personal tourism expenditure by province and territory. It also supplies information that identifies the reasons for visiting a region, the duration of stay and the activities undertaken while at the tourist location. The survey is a monthly supplement to the Labour Force Survey (LFS). The CTS provides totals for tourism expenditure, within a given region, as well as inter-provincially (inter-provincial exports and imports). The CTS, however, does not provide travel origin data for the territories, only travel destinations are available. Information from Statistics Canada's Survey of Household Spending (SHS) along with Input-Output supply data, are used to fill this gap. Estimates for pre-trip expenses, a portion of domestic demand, rely on the supply and disposition method using Input-Output data.
The ITS provides estimates for non-resident demand (separately for the U.S. and other countries) within Canada as well as imports of tourism (Canadian spending abroad). This survey was initially conducted to provide data for Canada's Balance of Payments with other countries.
Estimates for tourism supply are derived from Statistics Canada's Input-Output tables; as are Gross Domestic Product (GDP) and its components, labour income, mixed income and operating surplus. The Input-Output (I-O) tables are constructed using several large matrices of data that record the inputs (what is needed to make a good or service) and output (the goods and services provided). The output table shows the supply of various commodities by industry and therefore tourism supply can be calculated using these data. The I-O tables also provide a matrix of value added or GDP. Employment data come from the Canadian Productivity Accounts of the CSNA. These Accounts provide information on employment following SNA principles and using I-O industries. At the aggregate level, the number of jobs in this database is benchmarked to the Labour Force Survey (LFS). The industry distribution of these jobs, however, is primarily based on information from the Survey of Employment, Payrolls and Hours, although other industry survey and administrative sources are used as well.
Methodology
The goal of the CTSA is to measure the economic activity of tourism, including tourism GDP, employment, demand and supply. To do this, the CTSA takes demand data from the travel surveys (i.e. the CTS and the ITS) and calculates its contribution to GDP and employment. However, the source data for tourism are dissimilar in that some are based on commodity detail while others are based on industry detail. That is, demand data from the CTS and ITS are built up using commodity details while GDP and employment data use industry breakdowns. Thus, a link needs to be made between commodities and industries. Supply, which can be calculated using either commodity or industry detail, is the linking factor between these variables. A demand/supply ratio, as discussed below, can be calculated for each commodity. This ratio is then moved into the industry framework so that tourism GDP and employment can be estimated using the information provided by the demand surveys.
Supply
The starting point for the calculation of supply is the I-O tables, which give a detailed accounting of all the industries in the economy and the commodities they supply. The first step is to discern the industries that supply tourism commodities to visitors (see Appendix C for a complete list of tourism industries). However, the industry categories provided in the published Input-Output tables are not detailed enough for the CTSA.2 For example, the I-O tables display industry estimates only for the total food and beverage industry. The CTSA needs to split this industry, using information from I-O detailed calculations, into sub-industries since some of these sub-industries may not be included in the tourism account. For example, two sub-industries within the food and beverage industry are take-out food restaurants and caterers. Take-out food restaurants are considered a tourism industry and their supply is included in the estimate. Caterers are omitted since they are a non-tourism industry. Similarly, only tourism commodities remain in the calculation of total supply. In other words, only those commodities in the take- out food industry that are purchased by tourists are retained (e.g. meals) while non-tourism commodities are taken out (e.g. royalties).
After the tourism sub-industries and tourism commodities have been identified, supply shares are calculated. The supply shares are equal to a sub-industry's portion of the total output for a commodity within an industry. For example, if two sub-industries each provided one half of an industry's supply of meals, then the supply shares would equal 50% for these commodities for each sub-industry. These shares are used later to allocate demand by industry. It should be noted that taxes are added to all supply estimates obtained from the I-O tables to move the data from producer prices to purchaser prices, thereby matching the pricing used for the demand data.
Demand
The calculation for tourism demand in the CTSA begins with the survey data in the CTS and the ITS. The first step is to split the existing commodity detail provided in the surveys into the detail used in the I-O tables in order to allow appropriate demand / supply reconciliation. For some commodities, the existing survey data does provide appropriate detail. For example, information for the transportation commodities is sufficient to split these commodities into their I-O counterparts. For accommodation, information on nights spent in different types of accommodation is available from the survey and is used to split the commodities into the I-O classifications. Personal expenditure data from the National Income and Expenditure Accounts are used to split the remaining commodities into the I-O framework.
Once tourism demand is calculated using the same commodities as supply, supply shares are used to allocate the demand into sub-industries. As mentioned above, if 50% of meals is supplied by a sub-industry, this sub- industry is also allocated 50% of demand. After the demand data are distributed across industries, demand/supply ratios are calculated for each industry. These ratios are simply the total tourism demand of an industry divided by its total tourism supply. This ratio indicates what portion of an industry's output is attributed to tourism activity. Thus, if an industry demand / supply ratio equals 40%, this share of the industry's total output comes from tourism. These demand/supply ratios are later used to calculate tourism GDP and employment.
To increase the analytical usefulness of the data, demand is split into domestic demand and non-resident demand (international demand or exports) using information from the CTS and ITS. Data are also calculated for international imports. The availability of both export and import data allows for the calculation of the tourism trade balance.
Several additions are made to the survey data to provide a more complete estimate of tourism demand (see Table B1). The first of these additions is to include the non-fare spending by Canadians on the Canadian leg of a trip destined outside the country (domestic portion of international trips). This spending is not included in the published CTS estimates. For example, in the case of a person flying to New York from Winnipeg through Toronto, the spending associated with the Winnipeg to Toronto portion of the trip is added to the published survey data. The second addition is to include the fares paid by Canadians on international trips made using Canadian carriers (canadian fares of international trips). Since this service is produced domestically, it is included in domestic demand.
Another calculation is made to include the domestic tourism spending of territorial residents who are excluded from the CTS. Supply (revenue) data are available for all the tourism commodities in the territorial I-O tables. International demand is derived from the ITS and inter-provincial demand (i.e., spending by residents of the ten provinces who visit the territories) is derived from the CTS. Domestic demand for the territories is calculated using data from the SHS, which provides estimates of spending on tourism commodities such as accommodation and food and beverages.3
Pre-trip expenses or spending made by a traveller before a trip but for the sole purpose of travelling is another important addition made. These expenses include motor homes, travel and tent trailers, luggage and travel sets, tents and camping equipment and sleeping bags. The commodities included in pre-trip expenses were calculated using a supply and disposition method. Total exports of these goods were subtracted from total supply (including imports) of these items to obtain total disposition for Canada.
A special calculation is made in the CTSA for travel agency commissions which are not specifically identifiable in the travel surveys. This entails removing a fraction of the spending on various tourism commodities (e.g., air fares and hotel bills) and reallocating it to spending on services of travel agents. Essentially the demand for travel agency services is equated to the supply, as this commodity is used almost exclusively by tourists. The supply is just the commissions on travel arrangements and tour packages, which in I-O serve as the measure of output of the travel arrangements industry.
A final adjustment to the data arises from the demand/supply reconciliation. As commodities are reconciled between demand and supply, the tourism commodity ratio for a given commodity is examined. This ratio is at times considered too high or too low. This is particularly true for commodities for which little spending occurred. These issues were treated on a case-by-case basis. For example, suppose that after adding all relevant information for air transportation (information from the two travel surveys, the domestic portion of international trips, and the Canadian fares of international trips), total air transportation demand is 80% of total supply. The non-tourism exclusions to air travel demand could not account for this 20% difference. Therefore, the 80% total is adjusted upward to reflect a more realistic tourism commodity ratio for air transportation. In some cases, tourism demand can exceed supply for a given commodity. The supply and demand estimates are then examined to see which is of higher statistical quality and an adjustment is made based on this information.
Gross domestic product
Tourism GDP and its components, labour income, mixed income and other operating surplus, are obtained from the input table of the Input-Output tables. GDP is again calculated by sub-industry, just like supply either through information obtained for the sub-industries or, if no information existed, allocated by supply shares. The tourism industry ratios are then used to calculate tourism GDP. In other words, if the tourism industry ratio for a given industry is 50%, one half of the industry's GDP is allocated to tourism. This method also allows for the calculation of tourism labour income, mixed income and operating surplus associated with tourism.
A special calculation is made to derive GDP for the "other industries". These "other industries" (such as retail trade) are defined as non-tourism industries, or industries that do not meet the criteria for tourism industries but obtain some benefit from it. In other words, they produce some commodities bought by tourists. These commodities include groceries, alcoholic beverages from stores, pre-trip expenses, motor vehicle parts and repair and motor vehicle fuel and some other miscellaneous commodities (e.g. toiletries). I-O personal expenditure data are used to distribute these values across I-O commodities. These data are then moved into an industry framework using shares of output for these commodities. The totals are then multiplied by input shares so that the GDP components can be isolated.
Employment
Like GDP, employment is calculated at the industry level. Therefore, the employment calculation uses the same tourism industry ratios as GDP to calculate the tourism portion of each industry. Thus, if the tourism industry ratio for an industry is 50%, half of this industry's employment is allocated to tourism. This calculation is done for all tourism-related industries and then the shares are summed to arrive at total tourism employment within Canada. As mentioned before, employment data by industry come from the Canadian Productivity Accounts of the CSNA. The employment data are available only at an industry level too aggregated for the CTSA. Wages and salaries, from the GDP calculation, are used to allocate employment across the sub-industries.
Notes :
- A summary of the sources and methods used in the CTSA is presented in this appendix. A comprehensive examination of data sources and methods used to compile the CTSA is presented in the Canadian Tourism Satellite Account Handbook, catalogue 13-604 no.52, December 2007
- Sometimes even at the Input-Output "W" or worksheet level (the most detailed level available for I-O data) the industry estimates are not detailed enough for use in the CTSA. In these cases, survey level data are used.
- The SHS is carried out in the territories every second year. However, this does not coincide with years in which the CTSA is compiled. Consequently, estimates from the SHS for the year prior to the TSA, in this case 2003, are projected forward using indicators from the I-O Tables.
- Date modified: