3.0 International review of agriculture censuses and survey programs

An international review examined the programs in Australia, England, France, Mexico, the Netherlands, New Zealand, Norway, Portugal, Sweden and the United States. These countries were selected as they have an agriculture industry common to developed countries, yet with enough diversity in their programs to be of potential interest.

3.1 International review of agriculture censuses

The international study revealed that developed countries place a great deal of importance on their agriculture statistics programs. Most developed (and many developing) countries conduct a CEAG once every five years, except for those countries belonging to the European Union (EU), which are required to conduct a CEAG every ten years, supplemented every three years with a comprehensive farm structure survey. Some EU countries, on the other hand, have developed such an extensive agriculture administrative program that they are able to conduct a CEAG annually using data from these programs.

Table 1 shows the frequency with which CEAGs are conducted in the G20 countries. Aside from the EU, the majority of G20 countries conduct a quinquennial CEAG.

Table 1: Frequency of Censuses of Agriculture in the G20

Country1	Frequency of CEAG (number of years)
Canada	5
USA	5
Mexico2	5
Australia	5
South Korea	5
Japan	5
India	5
Russia2	5
South Africa	5
Indonesia	10
China	10
EU: Germany	10
EU: Italy
EU: UK
EU: France
Turkey3	10
Argentina	Irregular
Brazil	Irregular
Saudi Arabia	Irregular
1. The 20th member of the G20 is the EU itself 2. The decision was recently made to conduct a CEAG every five years 3. Following EU guidelines although not yet an EU member

CEAGs are mandatory in all countries studied. Response rates to the CEAG in most countries were above 95%, mostly achieved with significant follow-up. In England, the response rate was 73% and in the US, 85%. The response rate to the 2011 Canadian CEAG was 95.9% and has remained relatively stable over the years.

The majority of countries reviewed, including Canada, provide options to respondents in terms of the mode of data collection for the CEAG. The Netherlands has achieved a high response rate using the Internet. This option was offered on the 2006 and 2011 CEAGs in Canada, but limited broadband availability in rural Canada restricted this mode of data collection as a viable alternative to paper questionnaires at the time.¹⁷

The availability of broadband internet connections in rural areas is likely to increase substantially by 2016. As such, the internet collection vehicle will be more actively promoted as a method of data collection for the 2016 CEAG.

3.2 International review of agriculture survey programs

In general, most developed countries have extensive agriculture statistics programs that require significant input from agricultural producers. Countries, such as those that are a part of the EU, have strong data provision requirements, which provide a source of administrative data. Major survey programs play a crucial role for policy impact assessment of the Common Agricultural Policy in the EU.¹⁸

International agriculture survey programs are extensive and rely heavily on respondent cooperation. In some countries, fewer surveys are conducted than in Canada, yet those surveys (especially the farm structure surveys in Europe) are much more comprehensive than the targeted, commodity-based approach Canada uses. Some countries, such as Norway and Sweden, have strong administrative programs, which translate into less survey response burden, but impose heavier administrative compliance requirements on producers.

The majority of the intercensal surveys are mandatory in England and France, whereas in other countries, such as the Netherlands, they are voluntary. The Canadian model falls closely in line with the survey programs in England and France in terms of the mandatory nature of the majority of the surveys.

With respect to financial data, the European countries have a long history of collecting data through the Farm Accountancy Data Network. Different agents are used to collect these data. In France and the Netherlands, accounting firms are hired whereas in England a consortium of researchers from universities and colleges are retained to conduct these surveys. Respondents are often provided business management information about their own farm in return for participating. This reciprocity is sufficient to generate good response rates.

Canada is at the forefront with respect to the use of tax data, along with Australia. The Canadian Agricultural Taxation Data Program has been publishing data for decades. The Agriculture Division has more recently been studying the feasibility of replacing the detailed revenue and expenses questions from both the FFS and CEAG with tax data.

With respect to the monitoring and controlling of response burden, the Australian Bureau of Statistics has a systematic approach. In 1997, a statistical clearing house was implemented that requires all business surveys in all federal departments to obtain approval prior to conducting a survey or adding questions to an existing survey. This procedure has been found to be effective in reducing response burden by keeping unnecessary questions from being added to surveys, by modifying other questions and by preventing yet other surveys from going to the field. Due to its success, the clearing house will be extended to cover social surveys in 2012.¹⁹ Statistics Canada had a similar clearing house strategy until 1991 when it was eliminated due to budget cuts.

3.3 International review of remote sensing in agriculture statistics

Remote sensing technology uses computer analysis of satellite images to estimate earth characteristics. Many countries use remote sensing for agriculture statistics applications. The most advanced leaders include the US, China, Brazil and Europe, covering very large agricultural areas similar to the Canadian context.

The main objective of using remote sensing data at the international level is to forecast and estimate crop yield, area and production, and to monitor crop and pasture conditions. No country has exclusively used satellite data to replace a census or a survey, but many have successfully used it to support their statistical programs. For example, in China remote sensing is used to monitor crop area change, crop yields, production and growth, drought and other agriculture-related information for five main crops. In Europe, it is used to monitor crop vegetation growth (seven crop types) and to provide annual crop production forecasts. In the US, remote sensing is used to construct area sample frames for statistical surveys, which helps improve their accuracy. Remote sensing data are also used to produce maps of crop areas by major crop type, which can be used as a source of crop information between surveys.

The Remote Sensing and Geospatial Analysis (RSGA) section at Statistics Canada has worked over the years on numerous cost-recovery projects. The nature of these projects has determined the direction and development of this program according to client needs. Some of the projects of the RSGA include the following:

• The Crop Condition Assessment Program (CCAP): a web application that displays cropland and pasture conditions. Crop conditions are established for the entire country based on satellite data. This application is updated on a weekly basis during the growing season and is used by the agriculture community, including governments, grain marketers, researchers and individual farmers, to detect and delineate areas under stress.
• The crop yield estimation model: an experimental crop yield model developed to produce a crop yield estimate for the current year for spring wheat, durum wheat, barley and canola in western Canada.
• Support to the CEAG and survey programs: for example, CCAP satellite images and map products were used during the collection period of the CEAG to actively manage respondent burden by averting mail and telephone follow-up in areas of natural disasters, such as the flooding in Manitoba in 2011.

Further potential exists to more fully utilize satellite and agro-meteorological data to produce accurate crop area, yield and production estimates.

3.4 Lessons learned from the review

The international review provided information about how different countries collect the agriculture data they require. Often, Canada falls in the median position in terms of response burden, investment in the collection of agriculture data, the mandatory nature of data collection, the frequency of the CEAG, and the depth and breadth of the intercensal survey program. However, Canada does stand out with respect to the use of taxation data to generate financial estimates. The Australian Bureau of Statistics is also pioneering work in this area and is working with Agriculture Division to share lessons learned.

The international review also demonstrated that there is room for Canada to increase its use of administrative data and remote sensing technologies (especially in producing crop area, yield and production estimates).

The information gathered from the international review, coupled with an understanding of each country's agricultural landscape and sociopolitical structure, enabled an assessment of some international features that could potentially be developed to transform the Canadian agriculture statistics program.

The features of interest include

• the modular, comprehensive intercensal survey program of England
• the use of administrative data of the Northern European countries
• Australia and Canada's goal of increasing the use of taxation data to replace financial questions on surveys and the CEAG
• the different approaches used to determine the target and survey population thresholds
• the various methods used to maintain the survey frame
• Australia's Statistical Clearing House strategy.

These features of interest were used to develop alternative options. One of the goals of evaluating the options was to determine whether any of these programs could be sufficient in the absence of a quinquennial CEAG. Therefore, although some of these countries do conduct a CEAG on a quinquennial basis, each of the options was examined with the assumption of a decennial CEAG. In the section that follows, the current Canadian program is presented as the Baseline Option, against which the alternative options were evaluated.