Your location: Home
Compendium on methods and tools to evaluate impacts of, and vulnerability and adaptation to, climate
CRAM (Canadian Regional Agriculture Model)
The Canadian Regional Agriculture Model (CRAM) is a sector (i.e. partial) equilibrium, static model
for Canadian agriculture written in General Algebraic Modeling System (GAMS, details available at
http://www.gams.com/). The model has approximately 2300
variables, and 1300 equations. It consists of a main file with small files that can be
“called” when running the model. The model is disaggregated across both commodities and
space (55 crop regions and 10 livestock regions).
The basic commodity coverage is grains and oilseeds, forage, beef, hogs, dairy and poultry
(horticulture is excluded). For crops, data in CRAM include: yield, areas, cost of production,
prices, quantity demanded, input requirements (price and quantity of fertilizers, chemicals and
fuels), repair costs, management practice (tillage, summerfallow etc.). On the livestock side, data
include: number of animals, input requirements (feed, forage, pasture), variable costs, prices of
live animals (steers, bulls, cows etc.), prices and quantity of the different types of meat (high
quality beef, low quality beef, veal, pork). Dairy (i.e. types of milk and usage, butter, cream,
etc.) and Poultry (chicken, turkeys, eggs.) data are also included in CRAM. Other inputs include
shipment cost of commodities to ports, and demand information (price, quantity, and elasticities),
population per province etc.
CRAM is a non-linear optimization model maximizing producer plus consumer surplus less transport
costs. A Positive Mathematical Programming (PMP) approach is applied to crops, hogs and beef, and
input demand, feed demand, crop area and livestock numbers. Through a calibration process, the
model duplicates the observed allocation of a fixed and allocable resource (land) by positioning an
unobserved marginal cost curve in a manner that ensures the classic conditions for constrained
profit maximization are obtained. The model is processed through MINOS, that is, a general purpose
nonlinear programming (NLP) solver. More recently, experiments for more complex analysis have been
done with CONOPT. Both solvers are available under license (http://www.gams.com/sales/sales.htm).
The model currently reflects the baseline conditions for 2006 and 2010. Work is underway to determine
conditions for a longer baseline. The current CRAM model is static and is not intended for dynamic
application, although theoretically this could be possible.
CRAM has a national coverage where there are agricultural activities in Canada (Newfoundland and
Labrador, Prince-Edward-Island, Nova Scotia, New Brunswick, Quebec, Ontario, Manitoba, Saskatchewan,
Alberta and British Columbia).
The key outputs are: Consumer and producer surplus (CPS) in dollar terms; crop areas; livestock
production; quantities demanded; exports and imports. The model also indicates shifts (or needs) in
land resources in optimizing the CPS, when analyzing the impact of a policy (or a shock). It also
generates information on net income changes between the baseline and the scenario analysed (policy).
Results are given at CRAM regions level, but can be aggregated to reflect provincial and national
The main sources for data are:
- Statistics Canada, for Census of Agriculture and various data;
- Agriculture and Agri-Food Canada, for the Medium Term Policy Baseline (internal document);
- USDA, for international data.
Other sources of data might be needed depending on the relevancy of the information.
Ease of Use
In general, medium level of effort is required and, with time, CRAM is easier to use. Because most of
the economists who use CRAM have the necessary background on microeconomics theory, and in
optimization theory, most of the effort is in learning GAMS language. A period of up to six months
would suffice to gain familiarity with the language and the CRAM model itself. Furthermore, because
70 % of the raw series are available electronically, the data transformation part of the work with
CRAM might require a medium level, or less, of effort.
See ease of use
There is no formal training available
A Pentium-based computer with a minimum of 512 MB is required. However, for more complex analysis,
1GB is definitely better. The operating system used is Windows XP SP2 (for GAMS and solver
requirement on various platforms, please see http://www.gams.com/solvers/index.htm).
Description of the CRAM structure can be found in some of the early publications (such as Webber et
al, 1986). CRAM contains information on updates made, however this doesn’t constitute a how-to
manual for the models. There are on-going efforts with the CRAM Team to develop a separate
documentation that could fill (partially) the role of a manual. GAMS documentation is available at
One of the first applications of CRAM was to look at the implications of the introduction of medium
quality wheat on the Prairies (1986). Since then it has been used to examine the impact of the 1985
U.S. Food Security Act on the Canadian Grains sector (1988) and the impact of direct government
assistance programs on the beef and hogs sectors (1989).
CRAM has been employed within AAFC to examine the implications of the Canada-U.S. Trade Agreement
(CUSTA), the Multilateral Trade Negotiations (MTN) (1990), changing the Western Grain
Transportation Act (WGTA) (1991 and 1994) and licensing BST for dairy cows (1990). It has been used
for the environmental assessment of the crop insurance program (1998) and return on research
investment studies for wheat (1995), potatoes (1996), hogs (1998), and forages (2000).
CRAM has been used to calculate greenhouse gas emissions of the agriculture sector (with the
Canadian Economic and Emissions Model for Agriculture, CEEMA, 2000-2004), to track the
environmental performance of the agriculture sector (through agri-environmental indicators), and to
assess the expanded use of biofuels (2000). Furthermore, CRAM is currently used in an on-going
analysis of climate change impacts and adaptation.
Contacts for Framework, Documentation, Technical Assistance
Chief, Agricultural & Environmental Policy Analysis, Agriculture and Agri-Food Canada, 930
Carling Avenue, room 689, Ottawa, Ontario, Canada K1A 0C9; Tel: 613.759.7196; e-mail: email@example.com.
The incremental costs are for GAMS program and solver licences purchases. For the most up-to-date
information on costs, please consult http://www.gams.com/sales/sales.htm.
Agriculture and Agri-Food Canada. 2000. Options Report – Reducing Greenhouse Gas Emissions
From Canadian Agriculture, Public Works and Government Services Canada. Publication 2028/E, Ottawa,
Agriculture and Ari-Food Canada. Agricultural Policies and Soil Degradation In Western Canada: An
Agro-ecological Economic Assessment (Reports 1-5 http://www2.agr.gc.ca).
Fox, G., G. Brinkman and G. Thomas. 1998. The Economic Benefits of Canadian Swine Research.
Agriculture and Agri-Food Canada, Ottawa, ON.
Giraldez, J., R.J. MacGregor, B. Junkins, R. Gill, I. Campbell, G.Wall, I. Shelton, G. Padbury and
B. Stephen. 1998. The Federal-Provincial Crop Insurance Program: An Integrated
Environmenal-Economic Assessment. Agriculture and Agri-Food Canada, Ottawa, ON.
Graham, J.D., B. Stennes, R.J. MacGregor, K. Meilke and G. Moschini. 1990. The Effects of Trade
Liberalization on Canadian Dairy and Poultry Sectors. Agriculture Canada, Working Paper Number 3,
Heigh, L. and B. Junkins. 2005. Quantitative Analysis of the Impact of Agricultural Management
Strategies on Environmental Indicators. Agriculture and Agri-Food Canada.
Horner, G.L., J. Corman, R.E. Howitt, C.A. Carter and R.J. MacGregor. 1992. The Canadian Regional
Agriculture Model – Structure, Operation and Development. Agriculture Canada, Technical
Huffman, E.. 2000. Soil cover by Crops and Residue, pp. 33-40 in T. McRae, C.A.S. Smith and L.J.
Gregorich (eds.), 2000 Environmental Sustainability of Canadian Agriculture: Report of the
Agri-Environmental Indicator Project. Agriculture and Agri-Food Canada, Ottawa, ON.
Klein, K.K., G. Fox, W.A. Kerr, S.N. Kulshreshtha and B. Stennes. 1991. Regional Implications of
Compensatory Freight Rates for Prairie Grains and Oilseeds. Agriculture Canada Working Paper,
Number 3, Ottawa.
Klein, K.K. and B. Freeze. 1995. Economics of Loss Avoidance Research on Wheat in Canada. Report
submitted to Research Branch, Agriculture and Agri-Food Canada, Ottawa, ON.
Klein, K.K., B. Freeze and A.M. Wallbuger. 1995. Economic Returns to Yield Increasing Research on
Wheat in Canada, Report submitted to Research Branch. Agriculture and Agri-Food Canada, Ottawa, ON.
Klein, K., D. Le Roy, A.Walburger, S. Clark, and E. Yiridoe. 2004. Revisions to Crop Module in the
Canadian Regional Agricultural Model (CRAM) - Final Report. Agriculture and Agri-Food Canada,
MacGregor, R.J. and J.D. Graham. 1988. The impact of Lower Grains and Oilseed Prices on
Canada’s Grain Sector: A Regional Programming Approach. Canadian Journal of Agricultural
Oxley, J., B. Junkins, C. Dauncy and R.J. MacGregor. 1996. The Economic Benefits of Public Potato
Research in Canada, Report submitted to Research Branch. Agriculture and Agri-Food Canada, Ottawa,
Stennes, B.K., R.R Barichello and J.D. Graham. 1990. Bovine Somatotropin and the Canadian Dairy
Industry: An Economic Analysis. Agriculture Canada Working Paper No 1-91, Ottawa, ON.
Thompsen, J.G., R. Ewing and K. Shukla. 2000. Documentation and Analysis of the Economic Benefits
of Public Research of Forages in Canada. Report submitted to Agriculture and Agri-Food Canada,
Webber, C.A., J.D. Graham and R.J. MacGregor. 1988. A Regional Analysis of Direct Government
Assistant Programs in Canada and Their impacts on the Beef and Hog Sectors, Agriculture Canada,
Working Paper No. 6/88, Ottawa, ON.
Webber, C.A., J.D. Graham and K.K. Klein. 1986. The Structure of CRAM: A Canadian Regional
Agricultural Model. Department of Agricultural Economics, University of British Columbia.