«Daniel Roberts and Shane McFadden Abstract This paper discusses the lean hog hedge margin (Hog Crush Margin). We develop an Excelbased calculator to ...»
Lean Hogs Hedge Model and
Crush Margin Calculator
Daniel Roberts and Shane McFadden
This paper discusses the lean hog hedge margin (Hog Crush Margin). We develop an Excelbased calculator to determine this margin and provide fundamental analysis in relation to historic
margins. We discuss and explain all aspects of the model and give an application of the model.
AgStar Scholar Research Project
Department of Applied Economics
University of Minnesota
I. PROBLEM STATEMENTOur motivation with this paper is to better understand the swine crush-margin and develop a tool that will take into account all costs of production, market assumptions, and provide producers and lenders with the desired end result – a margin that can give light to hedging opportunities. The U.S. hog industry had, prior to January 2010, experienced a period of 30 months in which 29 months had negative profits. As we develop this model we are witnessing the highest margins this summer in hog production in more than a half decade. Many pork producers went into survival mode but the ones that effectively managed their margins and took advantage of hedging opportunities fared better than producers that did not.
One of the main problems for pork producers is that they do not have access to or familiarity with margin management because they do not have an accurate tool that is customized to their operation. The objective of this tool is to provide an accurate assessment of the potential margins. An understanding of how the current margins relate to historical margins can also aid in making management/marketing decisions.
This tool can be used by pork producers of any size or type of operation, provided that they are producing a market weight hog. This tool is not applicable for farrow-to-wean operations; however they can (and should) use it in coordination with finishing subsidiaries. It can also be used by lenders to accurately assess margin and recommend hedging opportunities to key clients. This tool is equipped to update automatically throughout each trading day which allows constant margin management and day-to-day comparisons.
The decision to hedge the margin at any particular time is difficult. Lenders can relate to the fact that it is difficult to get producers to perform margin management, especially in light of variable production and feed costs. This tool provides the user with an end result being the accurate margin for that producer and a comparison to average margins on a monthly basis. This tool is not designed to tell a producer what to do and it cannot make any hedging decision for the producer.
Model Data and Resources Most of the data for the model is based on Chicago Board of Trade (CBOT) and Chicago Mercantile Exchange (CME) prices for corn, soybean meal, and lean hog contracts. This data is obtained through the website http://futures.tradingcharts.com and, like all free market 2 | P a g e information is delayed 15 minutes. Real-time data could not be used due to its high cost.
However, if a user had access to this service it could easily be implemented. This information is pulled into an Excel spreadsheet through a web query and is automatically updated throughout the trading day.
We use the corn price from a southern Minnesota cooperative, Crystal Valley Cooperative in Lake Crystal, MN. This allows for local basis information to be added to the model. Basis levels are highly variable for corn depending on location throughout the country.
This model will be used by AgStar in Mankato, MN, so the basis information will be highly relevant. The user can, however, choose to use the CBOT corn data. If that is done, they will need to estimate their own basis.
Basis information for lean hogs is obtained from 3-year averages developed by Iowa State University (ISU Current Crush Margins). Basis information for soybean meal is difficult/ impossible to find as it is not published by processors. However, we have put in a simple assumption that the nearby three months will have a narrower basis in comparison to more distant months. The user can easily change this basis assumption depending on their experiences with soybean meal suppliers.
Cost of production information is obtained from Iowa State University averages (ISU Current Crush Margins). These averages are the default values. This tool does not serve strictly as a cost of production tool. Jayson Koopmans, ABC Credit Team Leader for AgStar provided insight into key functions necessary and critical analysis. Real data in relation to feed composition is obtained from Roberts Farms Inc. of Madelia, MN and developed in cooperation with Agri-Nutrition Services of Shakopee, MN. The feed ration composition is developed as the most highly cost-effective model in relation to the farm’s needs.
II. THE MODEL This model is designed to project future profits of a farrow-to-finish hog operation by using Excel tools to establish probable future prices of soybean meal and lean hogs. This is done by using futures prices and utilizing historical basis information (cash price minus futures price in a given month) as well as basis information. Excel tabulates the basis projections on soybean meal by taking into account seasonal trends in close relation to the soybean market, and similarly projects the basis for lean hogs by using historical averages in the respective months. The corn 3 | P a g e price is determined by the local bid price. These calculations are then incorporated into the other costs of production, and ultimately determine the projected profit per hundredweight or per head of lean hogs.
The model is the beginning of what can potentially become an intricate tool. While the corn bids on this model come from Crystal Valley Cooperative, where the Roberts Farm does business, figures for corn prices could be expanded to include different elevators at the click of a button. It would also be very useful to have a way to automatically update corn contract prices for local elevators the same way our model automatically updates futures market prices. This would save time and decrease errors by the user. As far as the assumptions and fixed costs, the model could also be specific to certain hog sites, diets, and could also be used in comparison to the corn, soybean meal, and lean hog prices when transportation costs and other site-specific considerations are taken into account. Finally, a more thorough analysis of historical margins could be performed to improve percentile analysis of the hedge profit.
As stated earlier, this model uses input information pertaining to the Roberts operation, and certain tables, such as the cost of production tab, are categorized according to their financial statements and in accordance with Iowa State averages. However, this model can easily be applied to most hog operations in southern Minnesota and throughout the Midwest. Costs of production as well as local feed costs can easily be adjusted, and will be automatically incorporated into the hedging calculation.
Another reason why this model is easily convertible to other hog operations is because it is based on a per hundredweight basis. That type of calculation makes it possible to compare different farms with each other. The input is either reliably standardized or based on local feed costs to be inputted by the producer. The cost per head will often be different based on the scale of the operation, but making simple changes to the costs in the model will account for those differences. The user can simply input their desired market weight and it will change all assumptions within the model.
III. OVERVIEW OF SECTIONSBefore any projections can be made, the fixed costs of production must be taken into account by going to the Cost of Production tab. These costs cover every expense per head from farrow-to-finish minus the volatile costs of corn and soybean meal. There is a flat rate for the 4 | P a g e cost of raising a pig from farrow-to-weaning, which in our model is $35 per head (see Exhibit 1).
From weaning to finishing, this table summarizes all the costs of vitamins/minerals, feed processing and delivery, health, labor, facilities, utilities, transportation, as well as interest cost.
Exhibit 1. Cost of Production
This exhibit from the model breaks down the costs, which can be changed by the producer at the touch of a keypad, and automatically totals it and puts the cost into the calculator. In this scenario, the total cost per head minus corn and soybean meal is $75.48. The user can change any of the cells highlighted in gray to fit the financial information available to them. For example, the weaned pig cost can be highly variable and the user may wish to use a general convention such as 50% of the 5-month delayed lean hog futures.
This tab contains a command button macro that enables the user to return to the default values at any time. The other command button will clear all values with a single click. As stated earlier, the user can also change the desired weight and applicable premium price on this page.
The premium is the percent over base that the operation is projected to obtain, or historically has obtained, because of minimal sort loss. This can be seen in Exhibit 2.
With the other costs established, it is necessary to move to the Feed Composition tab to break down and make assumptions on feed consumption per head, from wean-to-finish. Exhibit 3 shows the breakdown in corn and soybean meal feeding based on the stage of growth. The default model (based on diets composed by Agri-Nutrition Services) has three phases at the nursery stage and six phases at the grower stage with corn consumption increasing in each stage and the soybean meal-to-corn ratio increasing during the nursery stage and then gradually decreasing in the growth stage.
Recalculations can be done whenever there is a change in ration at any given stage. The formulations end with the amount of total corn and soybean meal per head is converted into bushels and tons, respectively, in order for easy comparisons to futures and cash prices of those products. For operations with a different number of growth phases or otherwise configured differently, a slight rearrangement of the table may be necessary. However, this task should not be difficult for most people with experience using Microsoft Excel. New columns can be inserted or existing columns can be renamed. We opted to make this step less automatic in order to preserve ultimate customization. Feed rations are extremely variable, and users should be allowed to either accept the highly efficient assumptions that are default to the model, or make up any variety of feed ratio scenarios. Further automation in this step would almost certainly inhibit the functionality of this model.
6 | P a g e Using the futures market price is the fundamental role of this model. The ability to set a price for commodities in the future is a valuable resource for producers, and this tool is a way in which to enhance that resource. Using basis data from Iowa State University (ISU Current Crush Margins), our model has a month-by-month breakdown of the lean hog basis using the averages of the first half and second half of each month. This historical data is an effective way of projecting future prices. Using basis is better than a straight prediction of price because basis trends are much more consistent and no matter what the futures price is in any given month or year the cash price will be correlated with the basis. Exhibit 4 shows the historical trends of the lean hog basis.
Exhibit 4. Lean hog historical basis
The basis for lean hogs usually tightens up in the summer months and stretches out between October and April. This is directly correlated with the biological production cycle where typically there is less pork produced during the summer months due to fall breeding difficulties (biological) and reduced weights due to higher temperatures.
Factoring in the basis for soybean meal is done in a simple, reliable way. The three months nearest to the current month have a basis of -$5.00/ton. All months farther out than three months have a basis of -$10.00.
A feature of Excel which enables this tool to be effective is the use of data updating from the web. Separate tables consisting of the futures prices of corn and soybeans from the Chicago 7 | P a g e Board of Trade (CBOT) and lean hog futures from the Chicago Mercantile Exchange (CME).
Using the Excel tab “Data”= “Get External Data”= “From Web” (Exhibit 5), it is possible to obtain constantly updated data for these prices.
Exhibit 5. Importing data from web This feature will inform the user of all the trading months, as well as other information in addition to the current price as an additional reference for the user.
The web query is set to automatically update every 15 minutes; this can be adjusted by the user if necessary. It is important to note that the user must enable data connections every time they open the spreadsheet to enable this function to work. Other conventions were performed to enhance automation, such as formatting the query to not recognize the date format and some functional conversions of queried names to match other queried names.
The information up to this point is all used in the calculator which projects potential hedging profits on lean hogs. The total cost per hundredweight is formulated first. This is done by taking the local forward price from Crystal Valley Cooperative. If desired, the user has the option to input the futures price for corn with the appropriate button, and then manually input a basis in the appropriate column if needed.
Soybean meal is tabulated by the futures price for the respective month, and then added to the projected basis as described to determine the cash price. These prices are multiplied by the number of bushels of corn and tons of soybean meal per head in the Feed Composition tab. This 8 | P a g e gives the total feed cost per head. The other fixed costs are then added to give the total cost per head, and then converted into total cost per hundredweight based on a 200 lbs/head carcass weight.