7.4 Personal expenditure at current prices

7.34 The purpose of this guide is not so much to describe in detail the methodologies currently used as to give the reader an overview of them. This section, which deals with current price estimates, is divided in three parts: the first two parts present the data sources and methods of the national and provincial/territorial estimators using summary tables while the third discusses the sales tax estimates. The next section deals with personal expenditure at constant prices.

7.35 Estimates of personal expenditure at current prices are derived using national and provincial/territorial estimators, as described in paragraphs 7.30 and 7.31. The data sources and estimation methods usually used to calculate the estimators for the 130 series of goods and services are presented in paragraphs 7.49 to 7.108. We say "usually used" because the source data used for calculating some personal expenditure estimators are occasionally inconsistent, making it necessary to resort to alternative calculation methods. These statistical inconsistencies are identified in the analysis that is conducted in part to confront the different statistical signals available, in order to assess their quality.

7.36 It should be noted that it is difficult to systematically present the methods of calculating estimators for each of the 130 series of consumer goods and services in a unique and standardized format while avoiding duplication. Conversely, it is also difficult to group them into a limited set of identical methods that would each cover a large number of series. However, it is possible to define two general calculation approaches, which can themselves be divided into six methods. The first approach consists of calculating estimators for expenditure series on the basis of direct indicators, that is, indicators that are very closely related to the series to be estimated. By contrast, the second approach consists of using indirect indicators, which do not in themselves represent a purchase of consumer goods or services. By its nature, the second approach tends to produce estimates of lesser quality. The calculation methods for estimators resulting from these two approaches are as follows:

• The direct "current prices" method [M1] consists of using an estimator that is an expenditure on goods or services. This estimator could differ from the series to be estimated because, for example, it includes not only the expenditures of households but also those of businesses and governments. The estimator for spending on furniture, derived from the Quarterly Retail Commodity Survey, represents an example of this method.
• The direct "quantity times price" method [M2] applies to goods and services series for which there are direct indicators of quantities. Expenditure estimators are calculated based on a quantity indicator that is converted to nominal values using consumer prices. This method is used for personal expenditure estimators such as those for certain types of transportation as well as energy consumption.
• The indirect "current prices" method [M3] applies primarily to series for which there is no current information on households' purchases, but which are nevertheless closely linked to a related series or a given industry through its revenues and/or its payroll. In particular, this method is used to calculate estimators for expenditure on shoe or clothing repair.
• The indirect "quantity times price" method [M4] is used when there is no direct or indirect source of information at current prices and no direct indicator of quantities. It consists of converting an indirect indicator of volume into an estimator of expenditures by applying a consumer price index to it. Quarterly estimators of expenditure on taxi services, among others, are calculated using this method.
• The "time trends" method [M5] is used as a last resort, when no reliable estimator is available. In such cases, the volume indicator will be based on recent or historical trends at constant prices, and a related price index will be applied to it. This method is used for quarterly estimators of expenditure on hairstyling services, among others.
• The specific methods [M6] are used for selected series which usually represent a major expenditure by households and require a complex or specific analysis. The complexity of these methods is due to the fact that they consist, in most cases, of using several types of indicators and/or a combination of the methods described above. This is the case, for example, with estimators of personal expenditure on new and used motor vehicles, on tobacco products, on paid rent, on imputed rent, on insurance services, on financial services indirectly measured and on travel expenditures.

7.37 The use of one method rather than another is based primarily on the availability of specific data when the estimators are being calculated. Since the calculation of a large number of estimators for consumer goods and services is based on two main data sources, namely the Survey of Household Spending and the Quarterly Retail Commodity Survey, the following paragraphs will provide an overview of these surveys and how they are used.

Survey of Household Spending

7.38 The Survey of Household Spending¹ (SHS) is an annual survey conducted since 1997.² Its main purpose is to obtain detailed information on the spending habits of Canadian households for a variety of goods and services, extending from food to shelter and including expenditure on education and health care. Estimates derived from this survey are published for different aggregates, including those based on household type, household income and geographic region. The data reflect the value at market prices and are reported annually by province, whereas territorial statistics are available from 1997 to 1999 and every two years starting in 2001. The Survey of Household Spending covers all private households in Canada, although certain groups are excluded, such as persons living in residences for senior citizens or in long-term care facilities, members of the Canadian armed forces living in military camps, Canadian residents posted abroad and persons living on Indian reserves. The above-mentioned groups are, however, covered under personal expenditure.

7.39 Conceptually, household expenditures are very similar to personal expenditure on consumer goods and services, despite some differences between them. The Survey of Household Spending data also include various outlays such as personal income taxes, vehicle registration fees, premiums paid to a public health insurance plan and donations to non-profit organizations, which are excluded from personal expenditure because they are considered transfers. The Survey of Household Spending also includes the value of all purchases of second-hand goods, while personal expenditure represents the value of purchases less sales of used goods.³ On the other hand, certain expenditures such as imputed rents of homeowners are part of personal expenditure but are either not captured by the Survey of Household Spending, or are treated differently by it.

7.40 The Survey of Household Spending is a primary data source used to project personal expenditure on selected commodities for the year following the last benchmark year. The Survey of Household Spending classification of expenditures is therefore adapted to the one used in the CSNA. For most personal expenditure series, a direct mapping exists, while for others, data from other surveys or from the Input-Output Tables are used to allocate various consumer expenditures from the Survey of Household Spending. No mapping is established when the concepts differ significantly.

7.41 Annual estimators for the following goods and services are derived from the Survey of Household Spending:

• J101 - Food and non-alcoholic beverages;
• J158 - Domestic services;
• J159 - Child care, in the home;
• J160 - Child care, outside home;
• J161 - Laundry and dry cleaning;
• J163 - Pet care;
• J178 - Interurban bus;
• J181 - Taxis;
• J183 - Telecommunications;
• J184 - Postal and courier services;
• J187 - Other recreational services;
• J189 - Fees for education and training, other than university;
• J191 - Hairstyling for men and women;
• J192 - Other personal care;
• J209 - Legal, accounting and other services;
• J211 - Miscellaneous household services;
• J219 - Cable television and pay television;
• J221 - Parking;
• J222 - Driving lessons and membership in automobile associations;
• J223 - Motor vehicle renting;
• J224 - Cinemas;
• J225 - Photographic services.

7.42 Depending on the personal expenditure series, either the value of the estimator, its rate of change, its trend or its provincial/territorial distribution is used. To project estimates for the second year following benchmark years and to distribute estimates on a quarterly basis, other data sources and/or methods of calculating estimators are used. These are described in Table 7.7 and Table 7.8.

Quarterly Retail Commodity Survey

7.43 Estimators derived from the Survey of Household Spending are primarily used to project personal expenditure on services. However, they are also used for some goods, but only as a secondary indicator since they have certain deficiencies. For example, since purchases of durable goods are by definition infrequent, it is possible that the sample drawn might not be sufficiently representative and hence the estimates of expenditure on these goods may have a higher variance. This is one of the reasons why the estimators for approximately 40 series of personal expenditure on consumer goods are calculated using data from the Quarterly Retail Commodity Survey (QRCS). Moreover, business surveys have the advantage of covering a major portion of the consumer activities of a country or a given region with a limited sample. This is not generally possible with household surveys, due to respondant burden considerations and other constraints.

7.44 It should be noted that the use of QRCS results as estimators for personal expenditure on consumer goods also has some deficiencies. These commodity sales represent not only the purchases of individuals but also those of businesses (including other retailers) and governments. Moreover, retailers are not the only sources of supply for consumers. For example, consumers may also make purchases from wholesalers or buy goods via the Internet. Because of these deficiencies, the use of estimators derived from the QRCS implies that household purchases of specific goods grow at a rate similar to the retail sales of those goods. This assumption is acceptable for many goods, such as food and non-alcoholic beverages. However, for other commodities, such as hardware goods which are bought in larger proportions by businesses, this assumption may result in less reliable estimators.

7.45 Since the first quarter of 1998, the QRCS has collected information on the sales of approximately 120 detailed series of goods and services for 19 types of retail outlets (trade groups) in Canada.⁴ The survey is conducted as a supplement to the Monthly Retail Trade Survey (MRTS), which collects data on total retail sales.

7.46 As a first step, the QRCS commodity classification is adapted to the one used in the CSNA. The concordance between the two classifications is often direct, but some QRCS commodities must nevertheless be distributed among several CSNA series of goods. When QRCS results are not available to calculate estimators for a given period⁵, the sales of each commodity from a previous period for a given trade group are then expressed in the form of a ratio representing them as a proportion of total sales. These ratios are then applied to the total of monthly retail sales from the Monthly Retail Trade Survey for the same trade group for the period to be estimated. The estimators are then calculated by simply summing the sales of each good for all the trade groups. An example is provided in Appendix 7A.

7.47 Quarterly and annual estimators for the following products are derived from the QRCS:

• J001 - Furniture;
• J002 - Floor coverings;
• J004 - Refrigerators and freezers;
• J005 - Washers and dryers;
• J006 - Stoves, ranges and microwave ovens;
• J007 - Other major appliances;
• J008 - Small electrical appliances;
• J010 - Garden tools and equipment for outdoor maintenance;
• J012 - Used motor vehicles;
• J014 - Road and off-road recreational vehicles;
• J015 - Motor vehicle maintenance and repairs;
• J016 - Motor vehicle parts and accessories;
• J017 - Radios, sound systems and accessories;
• J018 - TV sets, video equipment and accessories;
• J019 - Boats, aircraft and accessories;
• J020 - Photographic and optical equipment;
• J021 - Sporting and camping equipment;
• J023 - Musical instruments and supplies;
• J024 - Trailers;
• J026 - Recreation equipment rentals;
• J027 - Watches and jewellery;
• J051 - Men's and boys' clothing;
• J052 - Women's, girls' and infant's clothing;
• J053 - Thread, yarn and sewing accessories;
• J054 - Piece goods;
• J056 - Footwear;
• J058 - Luggage, leather goods and other personal effects;
• J059 - Toys, games and hobby supplies;
• J060 - Films and other photographic supplies;
• J061 - Household textiles and furnishings;
• J062 - China, glassware and kitchenware;
• J063 - Lamps, lighting equipment and accessories;
• J064 - Flatware;
• J065 - Hardware;
• J066 - Newspapers, books, magazines and stationery;
• J068 - Pets and supplies;
• J101 - Food and non-alcoholic beverages;⁶
• J103 - Pet food;
• J109 - Soaps and other cleaning supplies;
• J110 - Other household supplies;
• J111 - Pharmaceutical products and medical goods;
• J113 - Flowers, plants and other horticultural supplies;
• J114 - Cosmetics and toiletries

Other data sources

7.48 While the Survey of Household Spending and the QRCS are the main information sources for most personal expenditure series, several other data sources are also used to estimate selected series. These other sources include a great number of Statistics Canada surveys collecting information on the operating revenues of companies doing business in sectors as varied as transportation, accommodation, food services, recreational services and personal care. Other surveys, directly focusing on households, are also an extremely important source of information for measuring expenditure on other types of goods and services. These include the International Travel Survey and the Travel Survey of Residents of Canada. Information on prices, primarily drawn from the Consumer Price Index, is also used when the only data available are physical quantities. Lastly, a sizable number of statistics obtained from outside Statistics Canada, more specifically from administrative data sources and from various government agencies, provide valuable information.

National estimates at current prices

7.49 Table 7.7 provides a summary of the data sources and estimation methods usually used to calculate each of the national estimators at current prices, on an annual as well as on a quarterly basis. The majority of the estimators of personal expenditure are calculated using one of the first five methods presented in paragraph 7.36. These methods are referred to in this table as M1 for the first method, M2 for the second, etc. The table presents the estimation methods and data sources of the annual estimators for the two most recent years as well as those for the quarterly estimators. Because the calculation methods for some estimators are more complex than others (method M6), special attention is given to most of them in the following paragraphs.

New cars, vans and trucks

Annual method⁷

7.50 Personal expenditure on new cars, vans and trucks⁸ are estimated using monthly data from Statistics Canada's New Motor Vehicle Sales Survey (NMVSS) along with the monthly reports of the Canadian Vehicle Manufacturers' Association (CVMA). The survey presents the value and number of new motor vehicles sold by each manufacturer, while the reports provide the number of vehicles sold in Canada by model, in addition to identifying fleet sales. In the Canadian System of National Accounts (CSNA), the personal expenditures on new motor vehicles are expressed in terms of the full value of the vehicles that are purchased or leased by households; in other words, there is no consideration of trade-ins and of the tax savings resulting from trading used vehicles in the expenditure series on new motor vehicles.⁹ It should be noted that financial leases are treated as sales,¹⁰ although the applicable taxes are mostly distributed over the duration of the leasing contract on the basis of the lessee's periodic payments.

7.51 The data reported in the New Motor Vehicle Sales Survey are matched and confronted with those in the Canadian Vehicle Manufacturers' Association report. This reconciliation consists of applying the manufacturer's suggested retail price (MSRP) to the number of units of each model that are sold, after adjusting the Manufacturer's Suggested Retail Price to reflect rebates offered by dealers and manufacturers, transportation costs and the accessories commonly purchased as options.¹¹ The values thus obtained by model are then aggregated by manufacturer so that they can be confronted with the New Motor Vehicle Sales Survey data.

7.52 Once this reconciliation is carried out, the personal portion of new motor vehicle sales is calculated in two stages.¹² The first stage consists of classifying vehicles sold according to their main use, i.e., for personal or commercial use. Based on provincial and territorial vehicle registration data and Canadian Vehicle Manufacturers' Association statistics, this stage consists of subtracting from total sales of new vehicles all those clearly intended for commercial use, such as heavy-duty trucks and fleet sales, as well as all other new vehicles registered by businesses and governments. Although this stage serves to identify the purchases of individuals, it is not necessarily an estimator of personal expenditure, since some individuals, such as a number of self-employed workers, use their vehicle partly for commercial purposes. Conversely, some vehicles purchased by the business and government sectors are partially used for personal purposes. Since personal expenditure must represent the personal use that will be made of new motor vehicles purchased, the second stage will consist of subtracting from sales to individuals that portion representing the commercial use that will be made of the vehicles sold and adding a value reflecting the personal use that will be made of vehicles sold to businesses and governments. In practice, reflecting the personal use of all vehicle sales is a difficult task. Nevertheless, adjustments are derived on the basis of commercial use ratios drawn from the Survey of Household Spending (Survey of Household Spending) as well as data from the T4 Supplementary file of the Canada Revenue Agency, reflecting the personal use of purchases by businesses and governments. According to all of these calculations, personal expenditure generally accounts for approximately 70 percent of the value of new car sales and 50 percent of the value of new van and truck sales.

7.53 Apart from financial leases, where taxes are based on periodic payments, the calculation of federal and provincial sales taxes is based on the price of new vehicles, not accounting for trades that are made. The effect of these trades on taxes is discussed in paragraph 7.65.

Quarterly method

7.54 As in the case of the annual method, quarterly estimates of personal expenditure on new cars, vans and trucks are calculated using monthly New Motor Vehicle Sales Survey (NMVSS) data and the monthly Canadian Vehicle Manufacturers' Association (CVMA) reports. However, unlike with the annual method, the calculations use data by vehicle model aggregated into selected groupings.

7.55 The first operation in deriving the estimator of personal expenditure on new cars consists of calculating the proportion of car sales that are not fleet sales in relation to the total number of units sold and applying this proportion to the value of total sales (that is, without distinction as to the model or manufacturer) reported in the New Motor Vehicle Sales Survey. This estimated value of new cars purchased by individuals is then adjusted to reflect the personal use based on information derived from the annual methodology. The value representing the commercial use is included under business gross fixed capital formation on machinery and equipment.

7.56 The method of calculating the estimator of personal expenditure on new vans and trucks is similar to the annual method, except that sales by model are grouped into sales by vehicle category. Seven categories have been identified: light, standard, sport-utility, van, medium-duty, heavy-duty and bus. The first operation consists of applying average prices to the number of units sold, by category. A reconciliation between the values thus estimated and the official estimate from the New Motor Vehicle Sales Survey is then performed by manufacturer, and an adjustment is made to the average price per vehicle category for each manufacturer so that the difference can be allocated to them proportionally. The next operation is to subtract from the value of total sales of new vans and trucks all vehicles intended for commercial purposes, such as medium-duty trucks, heavy-duty trucks and buses as well as the value of fleet sales in the other four categories. This estimated value of new vans and trucks purchased by individuals is then adjusted to reflect the personal use based on information derived from the annual methodology. The value representing the commercial use is included under business gross fixed capital formation on machinery and equipment.

Used motor vehicles (net value)

Annual method

7.57 As in the case of other used goods purchased during a given period, only a portion of the value of used vehicle purchases is included in gross domestic product (GDP). For personal expenditure before taxes, this portion is equal to the value of purchases less sales of used motor vehicles made by households, hence the concept of net value of used motor vehicle purchases. Since the value of purchases and the value of sales for interpersonal transactions (i.e. no dealer involved) cancel each other out in the calculation, what is actually recorded is the net value of purchases made by households from other sectors of the economy.

7.58 Households have historically been, and still are, the main purchasers of used motor vehicles,¹³ and with the exception of interpersonal transactions, most expenditures of this type are made at new and used motor vehicle dealerships. Since these dealers are actually intermediaries between economic transactors wishing to purchase or sell used vehicles, the treatment of transactions carried out by households with dealers will depend on the sector of origin or destination of the vehicles involved in those transactions. Specifically, the net value of personal expenditure on used vehicles will include the total retail value of the sales made by dealers of vehicles that formerly belonged to businesses, governments or non-residents. However, only a portion of the value of sales to households of vehicles that formerly belonged to other households will be included in the calculation, since the ownership of these vehicles does not change sector. The part or portion that is considered in the latter case is equal to the profits generated by dealers on the sales of the vehicles in question as well as the expenses that they incurred, such as vehicle reconditioning costs, salaries of salespersons, etc. Conversely, the value of used cars, vans and trucks sold by dealers to other sectors will be deducted from the net portion of personal expenditure, if the vehicles previously belonged to households, either owned or leased.

7.59 Table 7.4 presents an example of the calculation of the net value of personal expenditure on used motor vehicles. Accounting for the fact that certain components shown in that table could change considerably from one year to another, the purpose of this example is to show the calculation method and to present the relative importance of each component. This method consists in the first instance of estimating the total margin generated by motor vehicle dealers on their sales of used vehicles, i.e. the profits and the incurred expenses excluding the acquisition costs, regardless of who the previous owners of those vehicles were or to whom they have been sold. Second, the dealers' acquisition costs of vehicles which previously belonged to businesses, to governments and to non-residents are added in order to reflect the total retail value of those vehicles that were acquired by the household sector. Finally, the purchases by other businesses as well as the exports of vehicles that previously belonged to households are subtracted in order to obtain the personal expenditure on used motor vehicles

7.60 The main data sources used to estimate the sales of used vehicles by motor vehicle dealers are the New Motor Vehicle Dealer Commodity Survey (NMVDCS), the sales of the used and recreational motor vehicle and parts dealers reported in the Quarterly Retail Commodity Survey (QRCS), and the federal Goods and Services Tax (GST) file. Despite its name, the New Motor Vehicle Dealer Commodity Survey is in fact a census carried out among motor vehicle manufacturers who answer on behalf of all their dealers. For its part, the GST file is a databank that includes, among other items, statistics on the revenues of companies collecting the GST on behalf of the federal government. The New Motor Vehicle Dealer Commodity Survey is used to calculate used vehicle sales made by new vehicle dealers, while the used vehicle dealers' market is estimated with data from the GST file and the QRCS.

7.61 Households' purchases of used motor vehicles from dealers are estimated by deducting from the dealers' total used vehicle sales a notional adjustment reflecting the purchases made by the other sectors of the economy, mostly unincorporated businesses. Because vehicle purchases made by households are treated differently depending on whether or not the vehicles previously belonged to other households, domestic sales and exports must be distributed by sector of origin and destination. In making these distributions, data on new vehicle registrations are used and various assumptions are made regarding the average duration of vehicle use before resale.

7.62 Profit margins realized by dealers on used vehicle sales are mainly based on data reported in the Guide d'évaluation Hebdo or the Canadian Red Book, while expenditures related to vehicle reconditioning are drawn from the New Motor Vehicle Dealer Commodity Survey.

7.63 The following example illustrates how used motor vehicle transactions involving an auto dealer affect personal expenditure and gross domestic product (GDP), using typical scenarios.

7.64 While this personal expenditure series records the net value of purchases made by households from other sectors of the economy, the taxes that are incorporated into those purchases are nevertheless calculated according to the gross value of used motor vehicle sales, including interpersonal sales.¹⁴  In fact, in addition to the taxes applicable to the gross value of used vehicles that are sold during a given period, two other components of taxes are also part of the calculation of this series, namely:

• the tax differential on trade-ins, and
• taxes paid on lease returns when purchased by lessees.

7.65 The tax differential on trade-ins consists of the sales taxes that a household does not have to pay on the purchase of a new vehicle when they trade their vehicle to a dealer. For a given transaction, the amount of taxes that a household will not have to pay is equal to the federal and provincial sales taxes applicable to the value of the used vehicle that will be traded in for a new vehicle. In the CSNA, the sales taxes on purchased vehicles reported in the new cars, vans and trucks expenditure series are calculated according to their full value. Therefore, by convention, the tax differential on trade-ins will be deducted from the net value of expenditure on used motor vehicles, in order to not overestimate the total value of sales taxes that is reported in the system.

7.66 It was previously noted that financial leases are treated in the CSNA as purchases of new vehicles, except that the taxes are mostly distributed over the duration of the contract on the basis of the lessee's periodic payments. Should the lessee of a vehicle decide to purchase it at the end of the contract, that transaction will not be recorded in the CSNA as a used vehicle sale, since this would amount to saying that the individual involved had purchased the same vehicle twice. This being said, in light of the convention that applies to the treatment of taxes on financial leases, the taxes applicable to the residual value of the vehicle the individual will pay at the end of the contract in order to take legal possession of it must be recorded in this series.

Quarterly method

7.67 The quarterly estimator of personal expenditure on used motor vehicles (net value) is calculated using the results of the Quarterly Retail Commodity Survey (QRCS) and the Monthly Retail Trade Survey (MRTS) as described in paragraph 7.43 and in Appendix 7A. This estimator represents the total retail value of used motor vehicle sales made by dealers, and its use assumes that its rate of growth is identical to that of personal expenditure on used vehicles.

Tobacco products

Annual method¹⁵

7.68 Personal expenditure on tobacco products requires a specific calculation methodology, which consists of estimating the quantities consumed, to which consumer unit prices are applied. A specific method is used because several sources of information are available and because personal expenditure on cigarettes and other tobacco products include not only legal purchases (that is, purchases for which all taxes were collected by manufacturers, distributors and merchants and were remitted to the administrative authorities concerned), but also expenditures attributable to contraband activities, making the estimation of this series relatively more complex.

7.69 The estimation of personal expenditure on tobacco products on an annual basis results from a reconciliation exercise between several data sources from both the supply side and the demand side. Among the main sources on the demand side are the revenues from specific taxes imposed on the various tobacco products by administrative authorities. These figures are drawn from the federal, provincial and territorial public accounts, as well as corresponding tax rates legislated by governments. Also noteworthy are data from household surveys conducted to evaluate Canadians' consumption habits: the Canadian Tobacco Use Monitoring Survey (CTUMS) and the Survey of Household Spending (SHS). On the supply side, the main data source consists of the domestic sales reported by the Production and Disposition of Tobacco Products survey, supplemented by declared international imports, obtained from the International Trade Division.

7.70 The first step of the methodology is calculating the quantity or volume of cigarettes, cigars and fine cut tobacco for which specific taxes have been collected by the federal, provincial and territorial authorities. This volume, which is expressed in terms of cigarettes and is obtained by dividing tax revenues by the specific rates levied in each province and territory, is then compared to data from household surveys and to the domestic sales reported by manufacturers, supplemented by international imports. Respondents to household surveys tend to underestimate their tobacco consumption, generally resulting in a marked difference between the volumes calculated from surveys and those derived from tax data. However, it has been found that in some provinces, especially those where specific taxes are the highest, the quantities obtained from tax data are below the reported consumption, a situation generally attributed to contraband. This would suggest that many smokers are turning to the black market for their supply of cigarettes and fine cut tobacco. This would also explain the fact that domestic sales reported by manufacturers have declined more substantially than the trends reported in household surveys each time major increases in specific tobacco taxation rates have been imposed in Canada.

7.71 The total volume of tobacco consumed in Canada (including contraband tobacco products) is calculated annually based on trends observed in household survey data, on the level of prices and on the domestic sales reported by manufacturers. Up to the period 2001-2002, the domestic sales reported by the manufacturers were generally a good estimator of the total volume of tobacco consumed in Canada, since the products of contraband were to a great extent composed of cigarettes produced by major manufacturers and purchased by illegal wholesalers for resale on the black market. With the federal and provincial tobacco tax increases of 2001 and 2002, this scheme was partly replaced as the wholesale prices of tobacco sold by the major manufacturers increased in accordance with the taxes. Since the contraband of cigarettes produced in unlicensed manufactures of tobacco products and the imports of illegal cigarettes from other countries took a greater share of the market, the domestic sales reported by the legal manufacturers of tobacco could not be used as a reliable estimator of domestic consumption anymore. In order to determine the total volume of tobacco products consumed in the country, information from a number of data sources is now used, including the trends observed in data from household surveys as well as the price-elasticity of demand for tobacco products.

7.72 Once estimated, the total volume consumed in Canada is then distributed by province or territory, using data drawn from the Canadian Tobacco Use Monitoring Survey, conducted by Statistics Canada for Health Canada since 1999.¹⁶ The consumption of contraband tobacco is then calculated residually for each province and territory by comparing, among other things, the total volume consumed to the quantity derived from tax sources on tobacco products.

7.73 Finally, consumer unit prices are applied to the volumes estimated by province and territory. The prices retained for the volumes sold on the legal market are drawn from the related consumer price index covering Canada's main population centres, while the prices paid for contraband tobacco are obtained from various sources, particularly media reports and press clippings.

Quarterly method

7.74 Since there is no sub-annual data source for the demand of tobacco products, the personal expenditure must be derived from the supply side. To approximate consumption, we use a moving average of the last four months of the volume of cigarettes sold in the country by Canadian manufacturers and by importers of tobacco products. The domestic sales of Canadian tobacco products are taken from the survey on the Production and Disposition of Tobacco Products by tobacco manufacturers with facilities in Canada. The import statistics are provided by the International Trade Division. Since taxes on tobacco products and prices vary from one province or territory to another, the national consumption of tobacco products must then be allocated among the provinces and territories using distributions from recent years as obtained by the annual method. Then, consumer unit prices are applied to the volumes estimated by province and territory. An adjustment is made to take expenditures on fine cut tobacco into account. Adjustments may also be made to reflect changing proportions sub-annually between consumption of legal and contraband products.

Paid and imputed rents

7.75 The annual and quarterly estimates of personal expenditure on paid and imputed rents are both derived using the same methodology. Additional information on this method and the one used to calculate estimates of net rent is presented in Chapter 5.

7.76 The estimation of paid and imputed rents begins with the housing stock, as measured by the Census of Population. The housing stock is divided into single-family dwellings, multiple dwellings, mobile homes, cottages, garages and farms. The Census figures on these stocks are extrapolated annually by the Investment and Capital Stock Division, which uses the number of new dwellings completed plus conversions¹⁷ less demolitions, from the survey of the Canada Mortgage and Housing Corporation (CMHC) and the Building Permits Survey. The housing stock is then divided between rented or owned dwellings and occupied or vacant dwellings.

7.77 The average rent is defined as the average price paid by renters for the use of a dwelling (single-family dwellings, multiple dwellings and mobile homes). This average rent is estimated using data drawn from part of the sample of respondents to the Labour Force Survey (LFS). The housing component of the Consumer Price Index (CPI) is also based on these same data.

7.78 The number of rented and occupied dwelling units (single-family dwellings, multiple dwellings and mobile homes) is multiplied by the average rent to obtain contract rent. The portion of contract rent that is not related to the dwelling space, that is, expenditures relating to facilities and services provided by landlords, is subtracted to obtain paid rent.¹⁸ These expenditures include depreciation of furniture, stoves, refrigerators and washing machines as well as costs related to water, electricity, heating, parking and maintenance services. Since the average rent is not available for cottages and farms, stocks and estimates of paid rent for other types of dwellings are used in order to obtain an approximation.¹⁹ For garages, the average rent comes from the Survey of Household Spending (SHS). For farms, data from the Census of Agriculture and from surveys conducted by Statistics Canada on farm operators are also used. Finally, an adjustment is made to eliminate the portion related to offices in the dwelling.

7.79 Several data sources are used to estimate expenditures relating to facilities and services provided by landlords, including the Survey of Household Spending, the personal expenditure on furniture and household appliances and on energy, and the Survey of Employment, Payrolls and Hours (SEPH) for maintenance services.

7.80 Data provided by Prices Division indicate that approximately 2% of workers use their dwelling as a workplace. The Income and Expenditure Accounts Division hypothesizes that 25% of rent should be attributed to the business portion. The adjustment for offices in the dwelling is calculated as follows:

Paid rent (before adjustment)
× Ratio of rent used for office space (25%)
× Ratio of individuals using their dwelling as a workplace (2%)
= Adjustment for offices in the dwelling.

7.81 Paid rent is the starting point for estimating rent imputed²⁰ to owner-occupants. By dividing paid rent by the rented and occupied housing stock, we obtain the average rent. This average rent is adjusted using a coefficient of quality, since a dwelling that is owned is generally larger and of better quality than a rented dwelling. The quality coefficient is based on the average number of rooms in owned dwellings compared to that of rented dwellings, according to the Census of Population. Imputed rent is obtained by multiplying the number of dwelling units owned and occupied (single-family dwellings, multiple dwellings and mobile homes) by the average rent paid and by the quality coefficient. The calculation is as follows:

Paid rent (J154)
÷ Rented and occupied housing stock
= Average rent paid
× Coefficient of quality related to a dwelling that is owner-occupied
× Owned and occupied housing stock
= Imputed rent (J153)

7.82 The imputed residential rent for garages, cottages and farms and the adjustment made for offices in the dwelling are derived similarly as those for rented and occupied dwellings.

Insurance services

7.83 The output of life and non-life insurance corporations represents the value of the service provided in arranging payments of claims and benefits in exchange for the receipts of premiums and contributions. Four insurance services are measured in personal expenditure: property insurance (J162), accident and sickness insurance (J171), auto insurance (J175) and life insurance (J202).

7.84 Personal expenditure for the first three insurance series is derived using the following formula:

(Premiums earned
- Claims due
+ Investment income on technical reserves)
× Personal ratio

7.85 Premiums are earned by insurance companies on a continuous basis over a period, although it is common for a premium to be paid at the start of the period. Premiums collected from policyholders are invested in financial or other assets which are held to meet future claims arising from the occurrence of the events specified in the insurance policies. These pre-paid premiums are a form of credit extended by the policyholder to the insurer and are part of the technical reserves. Similarly, claims become due for payment by the insurance company when the contingency specified in the policy occurs, but they may not be payable until some time later, often because of negociations regarding the settlement amounts. This is another form of credit described as reserves against outstanding claims and they are the other component of technical reserves. The investment income on these technical reserves accrues to policyholders and represents a premium supplement paid by policyholders to the insurer in the measurement of insurance output.

7.86 The measurement of life insurance output must also take into account the change in actuarial reserves. The actuarial reserves for life insurance are reserve requirements stipulated by the Superintendent of Financial Institutions and represent amounts set aside for payments of future claims and benefits which exceed the premiums and contributions received to the current date. For life insurance, the following calculation is made:

(Premiums/contributions earned
- Claims/benefits due
+ Investment income on technical reserves
- Increases (+ decreases) in actuarial reserves)
× Personal ratio

7.87 Data on premiums, claims and investment income for the provinces with a public automobile insurance plan ²¹ are obtained from the Public Institutions Division (PID). For the other provinces and territories, automobile insurance data are obtained from the Office of the Superintendent of Financial Institutions Canada (OSFI) and from the Quarterly Survey of Financial Statistics for Enterprises (QFS), compiled by the Industrial Organization and Finance Division (IOFD). For accident and sickness insurance and life insurance, data from the Canadian Life and Health Insurance (CLHI), OSFI and Industrial Organization and Finance Division are used, while property and casualty insurance is calculated using information from Industrial Organization and Finance Division and OSFI.

7.88 The personal proportions are based on historical data obtained from the OSFI. These ratios, used to determine the proportion of insurance output purchased by households, are as follows:

• Property and casualty insurance (70%)
• Accident and sickness insurance (70%)
• Auto insurance (70%)
• Life insurance (95%)

Financial services indirectly measured (FSIM)

7.89 In addition to explicit charges for selected services they provide, financial institutions also derive income from the interest rate spread between loans and deposits. This spread represents an implicit charge to borrowers and depositors for services that are provided by financial institutions without explicit fees. The following four personal expenditure series for financial services indirectly measured (FSIM) are calculated:

• Financial intermediaries, implicit loan charges (J200);
• Credit unions, implicit deposit charges (J201);
• Financial intermediaries, implicit deposit charges (J204) and
• Credit unions, implicit loan charges (J207).

7.90 FSIM estimates are calculated as interest received minus interest paid plus adjustments for own funds for all financial institutions that offer lending and/or borrowing services to households, except for credit unions. The calculation for credit unions is slightly different because these institutions are considered to be associations of individuals. Therefore, their profits are not part of corporation profits but are treated as redistributions to the persons and unincorporated business sector. The estimate of FSIM for credit unions is calculated as interest received minus interest paid plus adjustments for own funds minus profits of credit unions.

7.91 The FSIM estimates are roughly equal to the interest received by the financial institutions net of interest paid by them. The term roughly is used because the net interest received needs to be adjusted for the use of their own funds in carrying the lending and borrowing activities. The purpose of this adjustment is to insure that the net interest received (i.e. FSIM) actually reflects the service level provided independently of the source of funds. For example, a client of a bank should receive the same amount of service when borrowing money whether it comes from the bank's own funds or from its deposits. Similarly, the level of service received by a depositor should not be reduced when the bank pays interest on bonds issued. Basically, the own funds' adjustment corresponds to interest paid for non-deposit interest minus interest received for other interest bearing assets excluding mortgage and non-mortgage loans.

7.92 The FSIM estimates are then distributed into a depositors' portion and a borrowers' portion representing the service provided to both types of transactors. This distribution is carried out by type of financial institution using their assets and liabilities. For example, if a financial institution has $12 million in deposit liabilities and $8 million in loan assets, then 60% (i.e. (12/(12+8))) of the FSIM would be allocated to depositors while 40% would be allocated to borrowers.

7.93 The depositors' portion and the borrowers' portion are further divided among persons, governments, non-residents and businesses (corporations, government business enterprises and unincorporated businesses). The depositors' portion is broken down using the assets of each sector (or liabilities from the financial institution's perspective). The borrowers' portion is broken down using the liabilities of each sector (or assets from the financial institution's perspective).

7.94 Estimates of FSIM provided to persons appear in personal expenditure under the four series mentioned above while those allocated to governments and to non-residents are also included on the expenditure side of the production account under government current expenditure on goods and services and exports of services, respectively. FSIM estimates allocated to corporations, government business enterprises and unincorporated businesses are treated as intermediate expenses.

7.95 Source data for the chartered banks are obtained from the OSFI while data for the other institutions of the personal and commercial banking industry, for consumer lending and sales financing, and for credit unions are obtained from the Quarterly Survey of Financial Statistics for Enterprises and from other administrative data sources.

Net expenditure abroad

7.96 As noted in paragraph 7.8 and paragraph 7.26, while personal expenditure on consumer goods and services is defined as spending by Canadian residents at home and when temporarily abroad, the individual series (J001 to J214 and J219 to J226) do not reflect this concept.²² It is therefore necessary to add spending of Canadians abroad to personal expenditure and remove spending by non-residents (personal and business) in Canada from it. This is done through the aggregate net expenditure abroad, which adjusts the sum of the 127 individual series to reflect the total purchases of goods and services by Canadian residents wherever they were produced. Conceptually, travel payments abroad (J215)²³ and spending of military personnel abroad (J216) are personal expenditure, but they are not part of Canadian production. Offsetting entries are therefore made in imports in order to cancel these expenditures in the derivation of gross domestic product (GDP). Conversely, travel receipts from non-residents (J218) are subtracted in the aggregate from personal expenditure and are shown instead as exports, reflecting Canadian production.

7.97 The International Travel Survey (ITS) is the major data source for the calculation of net expenditure abroad. It provides data on travel spending and international transportation fares which are used to derive travel payments abroad and travel receipts from non-residents. Additional data on the education and health related travel spending is obtained from Culture, Tourism and the Centre of Education Statistics in order to adequately calculate J215, J218 and the personal expenditure aggregate.

7.98 The travel payments abroad series (J215) is the sum of two components. The first is derived by removing the travel spending of businesses as well as the spending of Canadians on education and health related travel from the total travel payments obtained from the International Travel Survey. The second component is the international transportation fares paid by households to foreign carriers which is calculated using a personal ratio of fares paid, derived from the International Travel Survey. The travel payments are recorded as imports of travel services, and the international fares as imports of transportation services.

7.99 Travel receipts from non-residents (J218) is also the sum of two components: the receipts for both personal and business travel spending from which the expenditures of non-residents on education and health related travel are removed, and the international transportation fares paid by non-residents to Canadian carriers. Those fares are included in J218 since they are already part of the individual personal expenditure series such as air transport (J179). The travel receipts are recorded as exports of travel services, and the international fares as exports of transportation services, both reflecting Canadian production.

7.100 The spending of military personnel abroad series (J216) is calculated by multiplying the salaries and wages paid to members of the Canadian military abroad as reported by Income and Expenditure Accounts Division, with a ratio reflecting the spending of military personnel outside Canada.

Provincial and territorial estimates at current prices

7.101 Provincial and territorial estimates of personal expenditure are only available annually and are based on the same classification of goods and services as are the national estimates (see Table 7.6). Whenever possible, the same data sources and methods employed at the national level are used to calculate the provincial and territorial estimates. Interprovincial travel expenditure, which is discussed in paragraphs 7.105 to 7.108, represents a notable exception.

7.102 Generally, the provincial and territorial estimates are derived by distributing the national estimates by province and territory. However, for certain expenditure series, such as personal spending on electricity, on natural gas, on paid rent and on imputed rent, the quality and timeliness of the statistical information make it possible to calculate provincial and territorial estimates and summing them to obtain national estimates.

7.103 For most goods purchased at the retail level, the provincial and territorial estimates are obtained with the method outlined in Appendix 7A: a provincial/territorial retail commodity breakdown is applied to the provincial and territorial retail trade sales by trade group. The estimation of spending on services by province and territory relies on more varied data sources, such as the Survey of Household Spending, surveys on transportation, accommodation, food services, recreational services and personal care as well as on data from administrative data sources.

7.104 Table 7.8 provides a summary of the data sources and estimation methods usually used to calculate each of the provincial/territorial estimators at current prices for the two most recent years. Series of personal expenditure without sales tax are derived with these estimators. Depending on the personal expenditure series and the data sources available, the estimators are used either to directly distribute the national benchmarks or to project the provincial and territorial benchmark data. In the latter case, the difference between the sum of the projected provincial and territorial estimates and the national estimate must be distributed by province and territory while minimizing the relative impact on the rate of change of each of the provincial and territorial series. As with the national estimators, the six methods presented in paragraph 7.36 are identified in the table.

Net expenditure abroad

7.105 In the provincial and territorial economic accounts, the aggregate net expenditure abroad includes an international component and an interprovincial component. The first corresponds exactly to the net expenditure abroad outlined in paragraphs 7.96 to 7.100 with the difference that the travel payments abroad are distributed by province or territory of residence and the travel receipts from non-residents are allocated to the province or territory visited. The second component is an estimate of interprovincial travel expenditure of Canadian residents. For the country as a whole, the net interprovincial travel expenditure is equal to zero because the sum of interprovincial travel payments equals the sum of interprovincial travel receipts.

7.106 The international component of travel payments abroad (J215) and travel receipts from non-residents (J218) is calculated using the same data sources and methods as presented in paragraphs 7.96 to 7.100, that is mostly from the International Travel Survey. The interprovincial component is derived using payments by residents of a province or territory while travelling in another province or territory for pleasure and other personal reasons. Data for the latter component are obtained from the Travel Survey of Residents of Canada (TSRC). An adjustment is also made to reflect other expenditures not covered by the Travel Survey of Residents of Canada such as outlays by residents of a province or territory who work in another province or territory.

7.107 As is the case for the international component, spending on education and health related travel is not included with the interprovincial component of J215 and J218 since it is already included with the individual personal expenditure series. For example, the university fees of a resident of Yukon studying in British Columbia are included with the series university fees (J188) and not with the series J215 of Yukon. Note that those fees are included with imports of services for Yukon and with exports of services for British Columbia.

7.108 By international convention, Canadian diplomatic and military personnel posted abroad are considered residents of Canada and their personal expenditure is shown under a region called "Outside Canada".

Estimates of sales taxes

7.109 The estimators, for which the data sources and estimation methods were presented in paragraphs 7.49 to 7.108, are used to derive personal expenditure estimates at current prices excluding the federal and provincial sales taxes.²⁴ The federal Goods and Services Tax (GST) and the provincial sales taxes must be added to obtain personal expenditure at market prices.

7.110 When it replaced the manufacturers' sales tax in January 1991, the GST had been set at a rate of 7 percent and remained at that level until July 1, 2006, when it was reduced to 6 percent. It was further reduced to 5 percent on January 1, 2008. The GST is charged on most goods and services that are sold and is similar to the value added taxes (VAT) that exist in many industrialized countries. However, because most businesses are allowed to claim input tax credits or refunds for the GST paid on their intermediate expenses, it is to a large extent a tax on final consumption.

7.111 A general sales tax is also levied by all provincial governments except Alberta.²⁵ A general sales tax is defined as a tax which is legislated by a provincial act and which applies to most goods and selected services purchased. Historically, provincial sales taxes (PST) have been applicable mostly to the purchase of tangible goods, but the tax base has been expanded in many provinces in recent years. In 1992, the Quebec government harmonized most of its tax base to that of the GST. In 1997, the governments of Newfoundland and Labrador, Nova Scotia and New Brunswick adhered to the Harmonized Sales Tax (HST) Agreement.²⁶ More recently, other provincial governments have decided to apply the general sales tax to additional products.

7.112 In the Canadian System of National Accounts (CSNA), sales taxes are calculated by province and by territory for each personal expenditure series. The tax rate in effect and a taxable proportion are applied to each personal expenditure series at current prices without sales tax. For any series, the taxable proportion corresponds to the percentage of the personal spending for which the sales tax (GST or PST) is applicable. The use of taxable proportions is necessary because of various exemptions that apply to selected goods and services as well as to specific purchasers. The method used to calculate the GST and PST is identical, except in Prince Edward Island and Quebec where the provincial sales tax rate applies to the personal expenditure estimates including the GST. As in the case for the estimates of personal expenditure at current prices excluding sales tax, the sum of the sales tax estimates by province and territory must correspond to the national estimates. An example of the calculation of the sales taxes applicable to furniture for Ontario is presented in Table 7.5.

Continue

• 7.5 Personal expenditure at constant prices

Back to

• 7.3 Data sources and estimation methods
• 7 Personal expenditure on consumer goods and services

---

Notes

1. For more information on the Survey of Household Spending, see the User Guide for the Survey of Household Spending, 2005 (62F0026MIE), available on the Statistics Canada website.

2. Prior to 1997, an occasional annual survey, Survey of Family Expenditures (FAMEX) collected information on household spending.

3. For more information on used goods, refer to paragraph 7.15 and to paragraphs 7.59 to 7.67 which deal specifically with used motor vehicles.

4. Previously, the distribution by commodity of sales by type of trade was based on the annual results of the retail commodity surveys of 1974 and 1989.

5. The data are usually available with a lag of one quarter.

6. As noted in paragraph 7.43, the Survey of Household Spending is the main source used when available, while for other periods the QRCS is used.

7. This methodology is also used to derive the annual benchmark estimates.

8. Includes pickup trucks and sport-utility vehicles.

9. Trade-ins and their tax implications are reflected in the personal expenditure series on used motor vehicles (net value). See paragraphs 7.57 to 7.67 for more details.

10. This treatment is similar to the practice followed by manufacturers.

11. Data on prices by model and data on the cost of options are drawn from the Canadian Red Book (ISSN 0045-527X) and the Guide d'évaluation Hebdo (ISBN 1488-3821).

12. The "non-personal" portion appears under gross fixed capital formation in machinery and equipment.

13. However, a significant number of fleet returns and off-lease vehicles were exported to the United States in the early 2000s.

14. Although the federal Goods and Services Tax (GST) and provincial sales taxes apply to used vehicle sales made by dealers, interpersonal transactions, on the other hand, are only subject to the provincial sales taxes.

15. This methodology is also used to derive the annual benchmark estimates.

16. The provincial and territorial distribution of expenditures on tobacco products, derived from the Canadian Tobacco Use Monitoring Survey, yields results comparable to the data reported by the Survey of Household Spending.

17. These are additional dwelling units created from formerly non-residential buildings or from other types of residential buildings.

18. Paid rent is often referred to as gross space paid rent or gross paid rent for space, emphasizing that this estimate measures only the cost of the space occupied.

19. Paid rent for cottages, garages and farms accounts for less than 3% of the total.

20. Imputed rent is often referred to as gross space imputed rent or gross imputed rent for space, emphasizing that this estimate measures only the cost of the space occupied.

21. Quebec, Manitoba, Saskatchewan and British Columbia.

22. A notable exception is the education and health related travel spending. The individual personal expenditure series already include such expenditures by Canadians abroad and exclude those of non-residents in Canada. For example, the university fees of Canadians studying abroad are included with the series university fees (J188).

23. The term "travel" in the series travel payments abroad (J215) is not restricted to pleasure travelling, but refers to all non-business trips by Canadians outside Canada.

24. A certain number of taxes are already incorporated into the personal expenditure series at current prices without sales tax, such as the federal excise duties on tobacco, the federal excise taxes applicable on the purchase of some new motor vehicles, the air travellers' security charge, and the excise taxes on gasoline, among others.

25. Although there is no general sales tax in Alberta or in the three territories, taxes are nevertheless levied on certain goods and services. For instance, the four governments levy tobacco taxes; there is a tax on accommodation services in Alberta; and the government of Yukon charges a tax on alcoholic beverages bought in stores.

26. In accordance with the HST Agreement, the federal government collects a 13 percent tax on the retail sales subject to the GST in each of the three participating provinces; between July 1, 2006 and January 1, 2008, the HST rate was set at 14 percent while it was set to 15 percent prior to July 1, 2006. From the amount collected, the federal government keeps the revenue attributed to the GST and redistributes the share corresponding to the 8 percent provincial tax to the participating provinces.