The Corn System Project: Defining an Economically and Ecologically Sustainable Commodity Corn System
25 January, 2000
Sustainability Institute
3 Linden Road
Hartland, VT 05048
Contact Information:
Philip Rice 802-436-1277 philrice@sustainer.org
Elizabeth Sawin 802-436-1277 bethsawin@sustainer.org
Each year close to nine billion bushels of corn are harvested from Midwestern farms. This astounding volume of grain (enough to cover a football field under a pile of corn more than four miles deep) is the source of a complex system of production, trading, processing, transport, and consumption that carries the product of Midwestern soils literally around the world, in forms ranging from steaks and chicken breasts to soft-drink sweeteners and fuel alcohol. This same process carries Midwestern soils and fertilizers into the Great Lakes and down the Mississippi to the "dead zone" of the Gulf of Mexico.
The corn commodity system has evolved under diverse forces, so that it would be barely recognizable to the great-grandparents of current farmers. Some of the changes have been positive, allowing more corn to be grown with less labor and giving farmers access to new markets. Yet these benefits have not come without costs. There has been a continuing decline in the number of farms in the region. And some of the new agricultural practices have had unintended consequences for the soil and water quality of the region and the health of its inhabitants, human and non-human.
Do improvements in farm efficiency necessarily come at the price of rural communities and their natural resources? What choices could the participants in the corn commodity system make to consolidate the gains of the past fifty years while reversing some of the undesirable trends? Could there be such a thing as a major grain commodity system that is sustainable both economically and environmentally?
In a world where farmers don't know if they will be farming five years from now, where traders must focus constantly on the short term fluctuations from which profits are carved, and where input suppliers must race each other to incorporate the most recent research findings into their product lines, it may seem a luxury to step away and look at the whole system and its behavior over the long term. But, as the players in the corn commodity system face the uncertainties of globalizing markets, changing government programs, and increased scrutiny for environmental impacts, a look at the whole system and how it responds to change could mean the difference between being at the mercy of the system and guiding it in positive directions.
The goal of Sustainability Institute's Corn System Project is to provide tools and resources to help the participants in the corn-based economy take a long-range look at their entire system Ð from producers through processors to consumers Ð with the purpose of finding ways to make the system function more positively for all its participants.
The following paper is a description of this project, our findings to date, the questions these findings raise, and the future direction of the project.
In the past fifty years, the productivity of commodity-based agriculture has increased dramatically, with yields for corn increasing almost four-fold (Figure 1). However, along with this increase in productivity per acre have come undesirable impacts on the land and people of farming regions. If you are looking for them, the symptoms of a troubled agriculture are all around us:
Figure 1
Source: USDA
G
G
G
These statements describe a system that is not succeeding in sustaining all the people who participate in it or the natural resources that the system itself depends upon. These examples are indicative of three troubling tendencies of the corn system (and other commodity-based systems as well). These three behaviors are:
Our purpose is to understand the reasons for these three tendencies and to discover ways to weaken their impact on the system.
Pressure on the environment stems largely from the types of production technology adopted over time. Among the most serious consequences of modern agricultural technology are pollution of groundwater with agricultural chemicals (Figure 2), loss of topsoil from erosion-prone lands, and accumulation of nitrogen-containing compounds in the Great Lakes, the Mississippi River and the Gulf of Mexico, with associated threats to aquatic life and the livelihood of fishing-based communities.
Figure 2
Source: IA Dept. of Natural Resources
In turn, farmer's technology decisions are influenced by the economic pressures affecting them, pressures that, despite ever increasing yields, remains consistently strong (Figure 3). In corn production, for at least the past twenty-five years, the net return to farmers has oscillated right around zero (note the red line in Figure 3). The producer's net income per acre becomes reliably greater than zero only after the injection of government subsidies into the system.
Figure 3
Source: USDA-ERS
This steady financial pressure on the producer is related to another unwelcome trend in commodity-based agriculture (Figure 4), the loss of medium sized farms. Depletion of the rural population then undermines previously thriving schools, churches, and small businesses.
Figure 4
Source: USDA-ERS
Regardless of their political perspective or their position on particular questions of economic or agricultural policy, most participants in the corn system can agree that these trends are disturbing. Even more disturbing is the future that these trends seem to be indicating -- a Midwestern landscape populated by a small number of very large farms, with the producers still on the edge of financial disaster and the land and water bearing the consequences of the farmer's struggle to keep producing more and more corn.
Most agricultural policies, both those in place and the alternatives promoted by various constituencies, are aimed at combating these undesired characteristics of commodity agricultural systems. From subsidies and bailouts, land set-asides, and marketing programs to taxes on agrochemical and the dredging of the Mississippi River, most of the players in the commodity corn system have suggestions for ways to make the system work better.
But problems that have persisted for fifty years or more while policies and political administrations have come and gone, tend to arise from deep, systemic causes. The solutions to these types of problems are neither obvious nor easy. If they were, they would have been discovered by now. Attempted solutions can result in new problems, as when a new technology successfully boosts yields only to have farmer's gains be eroded when a glut of grain depresses prices. Politically easy solutions do not seem to make a permanent dent in the problem, but it is impossible to muster the will to try more radical ideas, because there is no way to be "sure" the radical idea will help.
In situations like these, where complex, long-lasting problems appear to be too big for single constituencies to solve, a process known as system dynamics simulation modeling has proven useful to help stakeholders understand the underlying causes of the problems they struggle with and to work out together effective new ideas for improving the system's functioning.
The following section describes how systems modelers at Sustainability Institute are working with stakeholders in the commodity corn system.
We began in May 1998, first with a deep immersion into the data sets that describe the behavior of the system over time -- trends in price, yield, income, and costs. In addition to mapping out the ways the system behaved over time we also created a "system map" of the flows of corn from fields to elevators to feedlots and refining plants (see Appendix A). This map also shows the associated flows of money, and capital required for the corn to move along its value chain, and the environmental impacts associated with each step.
These data about the system over time and space led us to begin the creation of a simulation model to help us understand how the parts of the system are interconnected and how these interconnections bring forth its behavior. We began with the production sector, the farms that produce the corn. There were two reasons for this choice. Most of the environmental impacts of the system are clustered in the production sector. Most of the existing policies that try to shape the behavior of the system are focused there as well.
For the past year we have been meeting with farmers, input suppliers, policy makers, rural bankers, academics, extension workers, and NGO representatives. We have listened to their descriptions of how decisions are made, and we have turned their understanding of the system into the rigorous language of a computer simulation model. By working with participants knowledgeable about the system we have come to understand the chains of causality that link different parts of this system. Yields rise and production moves ahead of demand, so prices fall. Income injected in the system allows producers to bid up the price of land, rapidly increasing land rents and eventually debt burdens, and driving economically marginal operations out of business. And so forth.
This process of building shared understanding of the system is a joint effort, passing back and forth between modelers and the system's participants. We ask questions, listen to the answers and try our best to represent what we hear. Then we present the model's assumptions and behaviors to our network of advisors and ask: "Does this capture your experience of the system? Where does it fall short? What surprises you?" Based on their responses we re-examine our assumptions and equations and revise the model to represent the real system more closely.
These rounds of trial, presentation, and critique help build ownership of the model and confidence in the insights that arise out of the simulation runs and policy tests. The corn model would not exist without a great deal of time and thoughtfulness contributed by many people from within the corn system. Especially critical to this project have been people at the Institute for Agriculture and Trade Policy and individuals from within the Farmer Summit, a farmer-led group working to improve the agriculture system. A more complete list of the people who have contributed to the model and the ideas represented in it is shown in Appendix B.
Over the past year, we have had the chance to present this work to a range of audiences, from groups of farming families in living rooms to participants in farmer conventions and to university researchers. In general, the insights we have watching simulation runs play out under different policy scenarios are not surprising to the people living and working within this system. That is a satisfying indication that the model is capturing the essence of this system. Even more gratifying has been our observation that the model gives people a forum for analyzing and discussing the underlying causes of their system's behavior, where they may previously only have had a partial, intuitive understanding, or where they may have thought they were the only ones who saw it that way. Participants in the corn system are not surprised by the model results, but they are excited by them. The model seems to provide a way to articulate and explore the consequences of their deepest understanding of the system that controls their lives.
The simulation model reproduces many of the historical trends (Figure 5 and 6) in the system and responds to perturbations like droughts or new sources of demand in much the same way as the real system (Figure 7). This gives us confidence that the basic assumptions in the model are valid and courage to explore the causes of the problematic aspects of the system's behavior.
Figure 5
Figure 6
Figure 7
One of the key behavior modes we would like to understand is the trend toward lower and lower corn prices. The model reproduces this behavior by incorporating a very simple decision rule which has been validated by our farmer-advisors.
The model assumes that, in the struggle to maintain income in the face of falling prices, producers attempt to maximize their yield. They do this by adopting any new, yield-boosting technology as long as the anticipated income gain from yield increases is greater than the cost of the new technology. The simulation model assumes that technology suppliers carefully price new technologies so that, most of the time, new technologies will pass this cost-benefit test and be purchased. The result of this simple decision rule is that many farmers adopt yield-raising technologies. While higher yields have the potential to increase individual incomes, the net effect of many producers making the same choice is higher overall production, which tends to decrease price and therefore reduce incomes.
This system pattern is known as the "tragedy of the commons". The individually sensible decision, to attempt to achieve higher yields to improve income, has disastrous consequences for the whole system. Other careful observers of commodity-based agriculture have noted the importance of the tragedy of the commons behavior.
The existence of this dynamic comes as no surprise to producers, who recognize the pattern, but are nevertheless trapped by the situation. The usefulness of our approach lies in our ability to define the types of policies that might break the grip of the tragedy of the commons.
Some policies, such as creation of new sources of demand, or removing a percentage of acres from cultivation, create only a temporary improvement in corn price and farm incomes (Figure 8). These policies fail to remove the incentives that causes individual producers to go on adopting new technologies and raising yields.
Figure 8
Another class of policies has the potential to create lasting change in the system, but only by shifting the problem to another part of the system. For instance, government price support programs that guarantee a minimum price or programs that pay farmers to take more and more land out of production each year can stabilize corn price. However, the expense of supporting prices or idling land increases each year, because the underlying dynamic leading to ever-increasing production is also not addressed by this class of policies. In systems language this all-to-common result is called "shifting the burden to the intervener".
There are several policies that undermine the driving feedback loop of rising yields and falling prices and stabilize corn price without requiring more and more intervention each year. For example, agreements to limit production per acre or a transition to a style of production that focuses on lowering input costs rather than raising yields are two examples of the kinds of changes that could stabilize prices and incomes (Figure 9).
Figure 9
When summarizing the dynamics of this part of the system, we often say that "yield gains lead to price pains". Everyone in the system is aware of the tremendous yield increases and also the dramatically falling price. The great benefit of the model is that it forces us all to see how inextricably linked these two aspects of the system are. Policy attempts that overlook this linkage will never produce lasting change in the system because any gains they create will be overwhelmed as long as the system is driven by the tragedy of the commons.
The importance of understanding and addressing this dynamic is also underscored by the fact that the three problematic behavior modes we have focused on in this analysis are all exacerbated by this yield/price cycle. As shown in Figure 10, efforts to maximize yields put pressure on the environment. Low corn prices put pressure on producers, leading to a percentage of farmers quitting each year, and depleting rural communities of their populations and economic base.
Figure 10
In recent months we have begun speaking about the model to a wider audience, at farm conventions, in academic forums, and at policy strategy sessions. The enthusiasm that has greeted the model, and the tendency of people to think of ways to use the it, to identify other people who should see it, or to list new components that should be added IMMEDIATELY to it has strengthened our resolve to push this project forward. The reactions we have encountered so far convince us that people trying to improve the way the system works are in desperate need of exactly what our model can provide: information, educational methods, and policy testing tools.
In a paper like this one, it is difficult to describe what it is actually like to use the model, but we list below some of the reactions of people from within the corn system who have had an opportunity to see the model in action.
Initial laughter and joking about the model's implication that supply control would solve certain problems in the system moved to a more rapt attention and interest when the effectiveness of this type of policy over other current approaches is demonstrated in policy testing simulations. In some way these simulation runs represented a new hope or a more compelling argument for addressing a problem everyone in the room already knew about at an intuitive level.
The executive director of a sustainable agriculture institute saw the model in South Dakota. A month later, he brought two university department heads to see a presentation of the model and explore the potential for involving research staff from their departments in this project.
After seeing a presentation of the model a member of a corn growers group brought additional members to see a presentation of the model the next day (a Saturday). Their reaction: "When will it be finished and available?"
Again and again we have had the experience of people arriving for a presentation of the model because of the credibility of the farmer or neighbor who invited them. "I can only stay an hour" they say on the way in the door, but two and a half hours later the conversation is still going strong.
Several corn growers groups have seen the preview of the model and want to know when it will be available in complete form. They want it in time for the next election or the next farm bill.
A member of an association of farm management professionals wants to learn how to use the model to inform his own work, and he wants to introduce the model as a tool to his colleagues at their annual conference in July.
At a two-day strategic planning retreat of representatives of fifteen different agriculture policy groups, working with the model changed the mood of the room. Conversation moved from the level of details to the level of causal relationships. Even when the conversation shifted beyond the scope of the model, participants were paying careful attention to the interconnections and the ways that the system might react to policy changes. Cycles of cause and effect had entered into the conversation.
Our Midwest advisors on this project are leading us in three directions.
The first direction is to expand the scope of the model. The current version has a very limited representation of the sectors that buy, transport, trade, process and utilize corn. Yet, in the real system, these players are likely to have complex reactions to the sorts of policies we believe would improve the situation in the production sector. How would these other sectors respond to such policies? Would they have helpful or detrimental effects in these other sectors? (As a first pass, it seems that these sectors are benefited by ever-falling corn price and would react with their own counter-measures if price stopped falling.) To answer questions like these, we need to continue the iterative process of model development, working with advisors from these other sectors to extend the simulation model to represent the complete commodity chain. We also will expand the model help us understand the implications for the environment of various policy scenarios.
The second direction is to develop educational tools and materials based on the model and to work with partners within the system to disseminate these materials. A number of farmer organizations have indicated interest in using a version of the model to help educate their constituents and their elected representatives about the underlying causes of problematic behaviors in the system. In the next two to three years we intend to work with these partners to produce versions of the model that can be used for workshops and outreach and to train representatives from these groups to use the model as a teaching tool.
The third direction involves fine-tuning the model so that it can be a robust tool for policy analysis, and developing the contacts and networks that will be needed to make this tool effective in creating and implementing farm policy. Work in this direction will involve collecting and testing policy platforms from a variety of groups, and developing an interface for the model that will make it a useful tool for "live" workshop settings where a group can develop and test new policy ideas. The major emphasis in this direction is to develop this policy workbench tool so that it is a dependable and well-known resource for the work leading up to the 2002 Farm Bill.
To learn more contact:
Beth Sawin and Phil Rice
Sustainability Institute
13 Spencer Meadow
Hartland, VT 05048
802-436-1277
bethsawin@sustainer.org
philrice@sustainer.org
System map of the flows of corn from fields to elevators to feedlots and refining plants. (Not available.)
Partial listing of people who have contributed to the development of the corn model
| Name | Capacity | Affiliation | Interest |
| Margaret O'Dell | Project funding | Joyce Foundation | Farm economy, environmental policy |
| Kathryn Gilje | Project Facilitator; Senior Associate | Farmer Summit; IATP | Education/outreach |
| Jerry Perkins | Conventional Farmer- Corn/Soy, feeder cattle | Farmer Summit | Education/outreach |
| Sheila Ehrich | Conventional Farmer; Seed grower; | Farmer Summit | Education/outreach |
| Barry Dunn | Livestock Production Specialist | University of South Dakota- Brookings | Systems dynamics; livestock related issues |
| Gary Goldberg | Former Hog farmer, Oklahoma; | American Corn Growers Assoc. (ACGA), CEO | Policy |
| Keith Dittrich | Corn Farmer, Nebraska | ACGA, President | Policy |
| Dan McGuire | Policy Chairman | ACGA | Policy |
| David Senter | Legislative Representative | ACGA | Policy |
| Linda Reineke | National Farmers Organization | Director of Grain Marketing Department | Education |
| Mark Ritchie | President | Institute for Agricultural and Trade Policy (IATP) | Policy |
| Niel Ritchie | National Outreach | IATP | Policy |
| Dennis Keeney | Water Conservation | IATP, Leopold Center | Policy |
| Paul Porter | SW MN Experimental Station, Lamberton | UMN, Agronomy | Classroom education |
| Dick Levins | Agricultural Economics | UMN; Extension | Land, education |
| Stan Stevens | Agricultural Economics | UMN; Extension | Grain marketing strategies |
| Cornelia Flora | Rural Sociology | Iowa State University | Social Capital |
| Paul Lasley | Rural Sociology | Iowa State University | |
| William Heffernen | Rural Sociology | University of Missouri | effects of market concentration |
| Lee Mammen | District Sales Representative | Pioneer Hi-Bred | Education |
| Dick Thompson | Sustainable agricultural practices | Practical Farmers of Iowa | Education |
| Dan Specht | Organic agricultural practices | Practical Farmers of Iowa | Education |
| Paul Johnson | Farmer, Iowa | Former head of NRCS | Conservation/ education |
| Daryll Ray | Stochastic simulation modeling of farm policy | Agricultural Policy Analysis Center (APAC), UT- Knoxville | Effects of Agricultural Policy |
| Phil Ellingson | Conventional farmer, Corn/Soy, feeder cattle | Heart of Iowa Coop | Education |
| Steve Core | General Manager | Corn Plus Ethanol Plant | Education |
| John Molina | Director of Admissions, Professor | Soils Dept., UMN | Nitrogen cycle, Policy |
| Mike McCarvel | Conventional farmer, Corn/Soy | Minnesota Corn Growers | Member Education |
| Dan Perkins | Field man | SW MN Farm Management Association | Member Education |