Nate Bruce, Farm Business Management Specialist, nsbruce@udel.edu
A commonly heard question in farming circles around this time of the year is whether to store corn or soybeans. There are many arguments on whether to store one or the other. A Monte Carlo statistical model was developed to analyze the potential profitability of both options. Monte Carlo simulations are used to determine risk and uncertainty by randomly generating values from an observed value range and estimating what will most likely occur. Monte Carlo simulations are often used by economists to evaluate business decisions.
In this case, the Monte Carlo model is simulating the profitability or loss of storing corn or soybeans per bushel with each individual simulation representing a production and marketing year. The model assumes that 100% of what is yielded is stored. Production costs for corn and soybeans going back to 2007 from the University of Maryland were used in the analysis and converted into 2023 real dollars to account for inflation. Corn and soybean prices were collected from USDA NASS going back to 2007 and converted into 2023 real dollars. Yields and production costs are randomly distributed in each simulation. Storage costs are estimated as $0.40 per bushel based off responses from the recently distributed grain marketing survey. When the model randomly distributes yields, the storage cost is determined based on the simulated yield value ($0.40 x observed yield). Net returns for the marketing year are estimated in the model and the profit per bushel number is found from each simulation. There are instances the model randomly generates a profit per bushel that is extremely high or extremely low compared to what may occur in the real world. This occurs when the model is using either extremely high or low production costs and generating an extremely high or low yield. Combinations as such can create significantly high losses or profits, but statistically will almost never occur in the real world. To further reduce observations such as this, any profit per bushel greater then $2.00 or less then -$2.00 returns a profit per bushel value of zero. A plus or minus range of $2.00 was chosen because this isn’t completely out of the norm considering the market volatility experienced in the last couple of years and what could be achieved from storing grain.
The model was simulated 5000 times and resulted in the same conclusion every time: that there is a greater profit per bushel from storing corn than there is from storing soybeans. When the model is run, the values are randomly distributed so individual observations will vary, however this conclusion is always the same. Storing corn will always result in a profit of nearly $0.10 per bushel more than storing soybeans. There is greater net gain opportunity for storing corn as compared to storing soybeans. The potential profit per bushel from storing corn versus soybeans is given below:
Corn – $0.25 / bushel
Soybean – $0.15 / bushel
Distributions from a simulation are given in the charts below. Each bar represents either a profit or loss per bushel range. On the left axis is the number of observations the simulation model generated for each profit or loss per bushel range
This is a commonly discussed topic and the results from this model can help grain marketing decision makers. This model only evaluates the potential profitability of storing grain. All factors in the grain marketing plan should be considered before opting to store any commodity on the farm. This model and topic will be discussed in more detail at the 2025 Risk Management Conference.