An Assessment of the Existing Information on Agricultural Energy Use in Canada,
and New Data Collection Strategy. Volumes I and II. By Tulay
Yildirim, Tom Allen, Varghese Manaloor and Richard White. June 1995, Saskatoon,
The objectives of this report are to provide information on the available
data, to assess their limitations, and to identify gaps in the existing
information. There has been no comprehensive compilation of this
information in order to assess its quality and to determine overlaps and
gaps prior to this study.
The CAEEDAC staff spent several months searching
for the existing data on agricultural energy use in Canada. There is
a large body of data compiled by Statistics
Canada. There are also some data that could be collected from energy
suppliers, public utilities, provincial finance departments' records on tax
rebate programs, and provincial surveys conducted by various agencies. The
quality of available data on energy use in Canadian agriculture is, however,
presently not of sufficient quality to rigorously support any type of
comprehensive analysis, or to be used in monitoring the sector's contribution
to GHG emissions. The existing data
can be, and have been, used for less detailed analysis of agricultural energy
Limitations and deficiencies in the existing
databases should be assessed within the
context of the factors affecting energy use in agriculture. These factors
are technological change, relative prices, and technical efficiency at farm
level. Technological changes in agricultural production and in input supply
sectors have significant implications on energy use, and in input-output
ratios. For instance, research in crop production that results in increased
yields inadvertently reduce energy intensity in production, so does the technical
change in manufacturing sector through increased device energy efficiency.
Changes in relative input prices induce farmers to substitute away
from more expensive inputs, and also to adopt technologies which can save
the more expensive inputs. Changes in relative output prices, through
their impact on cropping decisions, also affect energy use in production.
Empirical evidence shows that different levels of output could be produced
with the same levels of inputs by following different methods of input
application, i.e., by attaining different levels of technical efficiency.
Producers can increase output levels by reorganizing resources at their
disposal, and by modifying the method of application. Technical efficiency
could be improved through education and extension programs on best practice
techniques of input application. Improvements in technical efficiency,
in turn, could result in a costless method of reducing agricultural energy
The synergistic nature of multiple inputs in agriculture, where device energy
efficiency is one of many factors that affects energy use, requires dynamic
analysis to assess changes in input mixes or cropping systems. Within
this context, the following limitations and deficiencies were identified
in the existing databases:
Data on actual quantities of energy used in agriculture, by energy types,
are very limited, and available data on these quantities are not very accurate.
This deficiency has been identified as a problem by several potential
users of data who were consulted.
In cases where there are accessible data on actual quantities of energy used,
corresponding data on other inputs, such as farm machinery, labour, etc.,
are not available. Without a complete set of data that is reported
consistently over time, the available data cannot be used for any comprehensive
analysis of energy use in agriculture.
Tillage practices have a significant impact on both direct and indirect energy
use, and on net carbon balances, yet detailed data on tillage practices are
very limited and extremely idiosyncratic. Economic analysis, environmental
studies, and engineering models all require this information.
Pick-up trucks are major users of energy. Data compiled from tax filers
information includes fuel used to run these trucks as a farm business expense.
There is a need for good information on the use of pick-up trucks, such as
the distance travelled and device fuel efficiency, in order to separate energy
used in production activities from energy used in transportation.
Data on farm machinery and equipment that
are used during a production run, which are essential to end-use analysis,
Indirect energy use is a significant part of total energy use in agriculture.
Yet data on indirect farm energy, i.e., fertilizers and other farm
chemicals, are available at only a very aggregate level. The quantities
of various fertilizers sold are reported by province, but the quantity data
on other farm chemicals are not available. In order to arrive at the
indirect energy content of these inputs, it is essential that fertilizers
are reported by type, since each type requires a different amount of energy
during the manufacturing process.
Farm labour and the opportunity cost of the farm operators time are very
important factors in determining technology choices, which, in turn, affect
farm energy use. The data on farm labour and wage rates are reported
as aggregate expenditures; better data are needed in order to analyze the
impacts of changes in farm labour and in off-farm job opportunities on technology
choices and agricultural energy use.
Some databases contain farm-level data; however, due to confidentiality
restrictions these data are not generally accessible. Farm-level data
are necessary for analyzing input-specific technical efficiency, which bears
useful policy implications by identifying the "best practice techniques"
and by indicating the inefficient use of inputs.
There is not sufficient information on farmers attitudes towards adopting
new technologies, in general, and energy-saving technologies, in particular.
Factors affecting technology adoption are very important in determining
the rate and speed at which energy-saving technologies may be adopted.
Clearly, a new database in agricultural energy use is needed to fill the
gaps in the existing data. The new data collection strategy should
be primarily responsive to the main objective of the
that is to reduce GHG emissions by improving
energy efficiency in all sectors. A set of data collected for a particular
purpose may not be suitable for all other uses. Yet, statistical data
sets are public goods and it is desirable to have a data set that benefits
the largest number of users. The new database should include the following
energy quantities and expenses, by energy type, i.e., gasoline, diesel,
electricity, natural gas, LPG, etc., and by farm type;
number of trucks, by size and by fuel type, and distance travelled by trucks;
number and use of self propelled combines and other mobile powered equipment,
by fuel type;
quantity of major grains dried on farms, by energy type;
number of irrigation systems, by fuel type, and irrigated area;
quantity of fertilizers, by types, used on farms, and fertilized area;
quantity of herbicides and pesticides used on farms;
data on farm labour; hours worked by the farm operator, family members, and
hired labour and their respective wage rates;
data on tillage practices, area cultivated with different techniques, and
on crop rotations;
data on summer-fallow practices.
In view of these data requirements CAEEDAC
recommends that small annual surveys to be conducted in order to elicit the
information listed above. We propose that survey sample should be the
same as other Statistics Canada surveys,
such as the National Farm Survey and Farm Financial Survey. These two
surveys provide data on financial characteristics of farms, land use, farm
production, machinery and equipment sales and purchases, and farm receipts
and expenses, which are essential to analyzing the different aspects of
agricultural energy use. Statistics
Canada has also a new pesticide survey that is currently being field
tested. Pesticide information could be obtained from this new survey
if the samples were chosen consistently.
Data on the number and use of farm machinery could be collected through the
Census of Agriculture. The Census of Agriculture has already a set
of questions regarding the number of selected machinery on farms. These
questions could be expanded to include the use of these machinery. In
this way, database would provide all variables required by the end-use models.
It is also possible to incorporate the questions on farm machinery
and equipment in Farm Financial Survey, which is an annual survey. However,
it is our opinion that machinery on farms do not change drastically from
year to year, hence collecting these data in census years would probably
provide enough information for those who work with end-use models.
If a new farm energy survey is designed in a way to allow for cross-referencing
with the existing surveys, the questionnaire could be kept shorter and overlaps
in survey questionnaires could be avoided, but a complete data set could
still be created. We believe this is the most cost-effective way of
creating a complete data set which could be utilized by researchers across