Methodology and data sources

Warning View the most recent version.

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.

3.1 Broad methods of measurement

The OECD Handbook suggests four broad methods for measuring the non-observed economy. They include:

  1. upper bound estimation;
  2. special surveys and other supplementary data;
  3. data confrontation and discrepancy analysis and;
  4. macro-model methods.

The upper bound method is used in Statistics Canada studies.Note1 As mentioned in the OECD Handbook, “the essence of the procedure is to consider systematically for each component of GDP the maximum possible amount of non-observed economy (NOE) activities and to total the results to obtain an upper bound. The procedure can be applied by any or all of the expenditure, income and production approaches.”Note2 An example of this would be child care services; given the number of children under the age of five, and the proportion of families using child care services, it is possible to calculate an upper bound estimate of spending on child care.

Another method entails the use of special surveys and other supplementary data. The OECD Handbook indicates that “these are surveys that are not part of the basic data collection programme. They can take a variety of forms, being special surveys of expenditure, income, labour, time use, and opinion surveys. They can be designed to target any or all of the NOE problem areas – underground, illegal, informal sector and household production for own use. They can be conducted by the national statistical office or by other agencies including, for example, sample audits by the tax authorities.”Note3 For instance, this study examines tax audit files and compares them to survey information and to original revenues and expenses reported in tax returns. One difficulty associated with this approach is that the different sources of information are not always comparable. Consequently, differences between them may simply be due to misreporting as opposed to UE.

A third method suggested in the OECD Handbook is data confrontation and discrepancy analysis. The OECD highlights the fact that confrontation of data from different sources is an integral part of national accounts compilation. Examples of possible data confrontation are also listed in the OECD handbook,Note4 and used in the regular production process of the CSNA:

  • Enterprise survey data versus taxation data; wages paid versus taxes raised; sales of goods and services due to value added tax versus value added tax raised; and production versus production related taxes;
  • Enterprise survey data measuring the production of commodities versus enterprise survey data measuring purchases of commodities; supply of goods and services versus the use of them;
  • Expenditure survey data versus retail trade survey data; household expenditures versus retail trade;
  • Expenditure survey data versus income or taxation data; household expenditures versus available income;
  • Enterprise survey data versus labour force survey data; use of labour versus supply of labour; turnover, value added, intermediate consumption, etc. versus the use of labour.

Macro-model methods are discussed in Chapter 12 of the OECD Handbook but are not recommended as accurate measures of the underground or non-observed economy. The OECD suggests that these methods do not properly measure underground activity and provide very divergent results depending on the assumptions used.Note5

3.2 Overview of the CSNA estimation methodology

This study uses each of the first three methods listed above. The upper bound method is used to create most of the UE estimates relating to household spending and construction. Supplementary data is used for alcohol and tobacco and to refine estimates where possible and provide information on industry allocation. Data confrontation provides further indicators of the industrial distribution of UE activity.

Before explaining the approach used to measure UE GDP, the three ways to measure GDP are summarized. First, the final expenditure approach consists of summing the final expenditures on goods and services (household and government final spending, business investment outlays, and net exports). Second, the income approach consists in summing all the factor incomes (compensation of employees, gross operating surplus, and mixed income) generated in the production process plus net taxes on products and on production. Third, the industry or value-added approach consists of measuring the total production of each industry and removing the cost of intermediate inputs. When summed across all industries this results in the GDP at basic prices, GDP at market prices is obtained by adding taxes less subsidies on products to the GDP at basic prices.

Estimates of UE activity in Canada are most easily measured using the first approach mentioned above, by examining final expenditures by the various sectors in the economy: households, corporations, government, non-profit institutions serving households and non-residents. It is assumed here that final expenditures of government and non-profit institutions serving households are not underreported. For the other sectors, most of the estimates use the upper bound approach (see Appendix C for more detail). In the household sector, specific methodologies are used for final consumption of tobacco, alcohol, tips, and rents while an upper bound approach is used for all other commodities purchased by households. In this case, the UE activity is referred to as skimming.

In gross fixed capital formationNote6, in the corporate sector, the upper bound method is also used. Residential construction is considered the only investment activity that is prone to underground transactions. In the non-resident sector, UE exports are based on upper bound assumptions and imports, a deduction in the calculation of GDP, are estimated using a lower bound approach. While there could be some UE activities in the components of final demand where the study assumes no such activity (e.g. non-profit organizations and government), the assumption is that it is negligible.

Income-based UE GDP is benchmarked to the expenditure-based UE GDP estimate, and allocated to wages, surplus or mixed income based on assumptions outlined in Appendix C. It is important to note that returns to labour in the underground economy are assumed to be all in the form of wages, that is, there are no additional employer costs (e.g., EI and CPP/QPP premiums, employer pension contributions, etc.). Finally, consistent with the assumption made for the expenditure-side estimates, no revenue is assumed to flow directly to government from UE activities.Note7 In other words, taxes collected and subsidies paid out are not hidden.

The industry-based UE GDP estimate is also benchmarked to the expenditure-based UE GDP estimate at the total level. For all the components that have a specific methodology in the GDP expenditure-based approach, the allocation by industry is done directly. This covers tobacco, alcohol, tips, construction, exports and imports. For the estimates of skimming, a confrontation approach using two different methods to allocate the UE activity by industry is used. First, the estimates of skimming (see explanation in Sections C.1.1.1 and C.3.2 in Appendix C) are allocated to the industry in which they are produced. Second, information from tax audit files from CRA is utilized to confront and adjust this distribution.

To summarize, several key assumptions are used throughout the study, namely:

  • small business are more likely to participate in UE activities than large businesses;
  • unincorporated businesses are more likely to engage in UE activities than incorporated businesses;
  • the non-profit and government sectors do not take part in underground activity;
  • corporate investment in machinery and equipment as well as intellectual property is not subject to underground activity.

Assumptions like these are necessary, given the limited amount of direct information available on underground transactions.  While debatable, they are considered reasonable for current purposes. If new information became available or more realistic assumptions were determined, they would be taken into account.

Appendix C explains in more detail the specific methodologies and assumptions used throughout the study.

3.3  Volume

With this study, a measure of the real underground economic activity (i.e., adjusted for inflation) is introduced. This is done using the same approach as used for the published real GDP, the Chain Fisher method.Note8 The Chain Fisher method is considered the best approach to estimate the growth in volume and prices. The reference year chosen was 2007, the same as for the official GDP, which facilitates the comparison with official estimates.

The volume component of UE activity is measured by dividing the nominal value of a given series by an appropriate price index. This is referred to as deflation. For the UE estimates, the deflation of a series (such as household spending on alcohol) occurs at the lowest possible level for which a representative price index can be found. The deflated series are then aggregated together to derive various totals.

Real UE gross domestic product (UE GDP) is estimated from 124 series. The nominal values in these series are derived as described in the preceding sections. The price indexes are adjusted to remove tax for the household spending categories, as tax is not paid on UE purchases. Prices for gross fixed capital formation and exports are already “without taxes”. The prices for imports are not adjusted because of a lack of information.

3.4  Data sources summary

Several data sources are used in this study. CSNA estimates, available in June 2013, were used extensively. Among them, the 2009 Input-Output tables were used to prepare all conversion matrixes relating industry information to household spending information. The Canadian productivity accounts were used to derive GDP by industry on a nominal basis. The income and expenditure accounts provided the time series for household spending and gross fixed capital formation.

Census of Population, Census technical reports and the Survey of Household Spending were used for households and dwelling estimates used in the estimation of rent and the analysis of UE spending on a household basis. The Survey of Household Spending was also used to validate skimming estimates for some categories of spending.

LCBO annual reports and information were used as a basis to estimate illegal alcohol.

The overall consumer price index (CPI) was used to derive the threshold (in terms of gross business income at constant prices) that define “small” businesses in this study. It was used in conjunction with tax databases maintained at Statistics Canada by the Tax Data Division.Note9 These databases pertain to unincorporated (from the T1 income tax form) and incorporated business (from the T2 income tax form) income tax returns. By using this tax information it was possible to obtain the operating revenues and operating expenses on a NAICS basis for businesses in Canada. These were used as the main input into the skimming estimates.

The Tax Data Division also provided a tax audit data file for incorporated businesses showing the original information provided on the tax return and the latest information after all audits by the CRA. This file was used to estimate reassessment rates by industry. The estimates derived from this file were used to obtain a better industry allocation of skimming.


  1. Gervais, Gylianne 1994, The Size of the Underground Economy in Canada, Statistics Canada, Catalogue no 13-603-MPE1994002. Ottawa, Ontario. And Statistics Canada (2011) “Estimating the Underground Economy in Canada, 1998-2008”.
  2. “Measuring the Non-Observed Economy: A Handbook” (Organisation for Economic  Co-operation and Development, International Monetary Fund, International Labour Organisation and CIS STAT (2002)). p. 54.
  3. Ibid., p. 59.
  4. Ibid., p. 51.
  5. Ibid., p. 187.
  6. This is the national accounts terminology for investment in physical assets.
  7. This does not exclude the possibility that UE activities could indirectly contribute to government income. In fact income from UE activities eventually finds its way back into the formal economy, and thus generates tax revenue.
  8. See Chapter 2 of “Guide to the Income and Expenditure Accounts.” Statistics Canada, Catalogue no.13-017-XWE.
  9. The use of the CPI to establish thresholds is described in Appendix C, Section C.1.1.1.
Report a problem on this page

Is something not working? Is there information outdated? Can't find what you're looking for?

Please contact us and let us know how we can help you.

Privacy notice

Date modified: