Journal of the NACAA
ISSN 2158-9429
Volume 11, Issue 1 - June, 2018

Whole-Farm Planning Models for Assessing Inter-Generational Transition

Rosburg, B., Program Educator, Iowa State University Extension and Outreach
Griffin, T.W. , Cropping Systems Economist, Kansas State University

ABSTRACT

Rosburg Farms is a crop farm in Northwest Iowa specializing in corn and soybean production. The oldest farm operator was looking toward retirement while the youngest generation was identifying an entry strategy into the operation. The overall goal of this research was to demonstrate how whole-farm planning models can be adapted to evaluate a potential intergenerational transition. Specific objectives were 1) to present how whole farm resources such as machinery, labor, and time constraints were collected and inventoried and 2) determine how additional rented land affects crop mix and machinery resource requirements. Data specific to Rosburg Farms were entered into an existing whole-farm linear programming model for regional- and farm-specific information. Farm-specific parameters included labor, machinery, crop rotations, and available acreage. Results indicated additional fieldwork could be completed with available labor and machinery. The process of adapting planning models is of interest to a wide spectrum of Extension personnel; and model results are applicable beyond the case study farm to other beginning farmers, farms anticipating intergenerational transition, and succession planning specialists.

 


Introduction

Rosburg Family Farms is a multi-generation crop farm specializing in corn and soybean production in Northwest Iowa. A family member was transitioning into retirement while the next (third) generation was identifying how they could enter the operation as a full-time farmer. Altogether, the family farm operation was comprised of three decision makers (Rosburg, 2017). The oldest Rosburg desired to reduce his direct involvement with day-to-day operations and his son was trying to make majority of decisions. The eldest Rosburg’s grandson has been a self-employed seed salesman and works on the farm during nights and weekends. The younger two rent land for their own crop production. Since the youngest does not own equipment, he pays the older two for custom hire of their machinery and labor to conduct field operations. A portion of the Rosburg Farms case study (Rosburg, 2017) is reported here.

The overall goal of Rosburg (2017) was to determine how to optimize machinery and acreage of Rosburg Farms. For this paper, specific objectives were 1) to present how whole-farm resources such as machinery, labor, and time constraints were collected and inventoried and 2) determine how additional rented land affects crop mix and machinery resource requirements. Specifically, the farm decision makers are interested in determining if an additional 384 acres could be farmed in a timely manner with existing equipment inventory. In order to achieve these goals, a whole-farm plan must be parameterized for regional and farm-specific information. Parameterization is the process where existing planning models are adapted for specific farm and regional information such as acreage, rotations, and weather probabilities. In this paper, the steps taken are presented so that the reader can perform similar analyses for other farms.                                                                                                        

Whole Farm Impact of Acreage Expansion

Like many Midwestern crop farms considering expansion, Rosburg Farms desired to know how whole-farm timeliness would be impacted by acreage expansion while constrained to use the same equipment. Decision makers also needed to understand how overall cash flow would change and if that income would be sufficient for the youngest Rosburg to farm full time. Linear programming (LP) methodology was chosen to evaluate how expanding farm acreage impacts whole farm timeliness and profitability. Specifically, the Purdue PC-LP Farm Plan program (Dobbins et al., 2006) was adapted for northwest Iowa.

Before the model could be adapted, information on existing farm equipment, crop rotations, tillage systems, labor resources, timeframe of equipment use, and current farm size was collected. A list of proposed changes that might occur to the crop operation as a comparison to the current situation was identified. Once farm-level data were collected, the model was parameterized and results analyzed using standard financial statements. Income and production budgets were compared to the possible changes that could happen with additional acreage. Production enterprise budgets are generally available from Extension professionals at respective Land Grant Universities; this study used Iowa State University’s Ag Decision Maker Crop Production Costs Budgets for Corn following Soybeans and Herbicide Tolerant Soybeans following Corn (herbicide tolerant) (Johanns, 2017) and Estimated Costs of Crop Production in Iowa- 2017 (Plastina, 2017). Specifically, the farm would need to generate at least $220,000 per year to avoid the necessity of off-farm employment.

 

Literature Review

Linear programming (LP) is a method to estimate an optimal solution to maximize an objective function (Dantzig, 1949). Tice (1973) developed an LP model to select the most profitable machinery size, crop combination, and tillage practices under land, labor, and fieldwork constraints. Tice found that one laborer with a four bottom plow and average number of field workdays could farm approximately 700 acres. This was compared to using three-bottom plow equipment and average number of field workdays which resulted in one farmer being able to farm 625 acres (Tice, 1973). Griffin et al. (2014) used a similar framework to estimate whole-farm costs of conducting on-farm experiments. Their objective function maximized whole-farm returns to land, unpaid labor, and machinery. While Griffin et al. (2014) demonstrated how linear programming could be used to compare different scenarios, limitations of the model were also noted. Limitations included omission of any random or stochastic properties such as the risk of input parameters that utilize exact values. In other words, model results were only as good as the data utilized. There has been a history of using LP models in an Extension framework. Griffin et al. (2010) reported that Purdue has been offering an LP modeling service for several decades. Given that the model is expensive to update requiring specialized skills to help farmers input data and interpret the results, it was not likely that the private sector would provide similar services (Griffin et al., 2010).

To ensure the likelihood of completing fieldwork in a timely manner, larger machinery could be necessary during planting or harvest times being affected by adverse weather. Not finishing fieldwork in a timely manner could lead to decreased yield production, i.e. yield penalties. Equipment sizing decisions must consider fieldwork probability and labor availability (Williams and Llewelyn, 2013). Mensing (2017) reported that farm management implications were evaluated in relation to fieldwork probability, especially machinery utilization. Profit-maximizing producers must manage machinery resources so they are not over-equipped but have adequate capacity to plant and harvest farm acreage within available suitable fieldwork times.

           

Data and Methods 

A combination of mathematical modeling and financial analysis were used in this study. Linear programming (LP) was the mathematical optimization technique chosen to determine the feasibility of farming full time by examining optimal usage of crop acres, labor, machinery, and resources (Boehlje and Eidman, 1984). To formulate the model, farm-specific and region-specific information were collected and entered into the model. Once baseline LP model was formulated for the current farm operation (Rosburg, 2017) and output matched the current activities, additional scenarios were evaluated against the baseline. Linear programming model results were evaluated in a series of financial analyses.

For the financial analysis, enterprise production budgets and cash flow statements were used to evaluate candidate scenario profitability. Enterprise budgets were created for soybean and corn production based on Iowa State University Ag Decision Maker Crop Production Budgets (Plastina, 2017). Custom rates from Iowa State University Ag Decision Maker, Custom Rate Survey were used to determine opportunity costs and machinery expenses (Plastina et al. 2016). An opportunity cost is a benefit that could have been received but was foregone to take another course of action. For example, machinery could be used for custom farming but was used for fieldwork on Rosburg Farms. Production budgets include expected costs and projected income from crop sales. Budgets allowed the farm decision makers to calculate projected income or loss to determine if it was financially feasible to rent the additional crop acres.

Farm-specific Constraints

At the farm level, information on crop rotations, machinery resources, and labor availability were collected. Laborers were available 6 to 7 days per week. During peak fieldwork times such as planting and harvesting, unpaid laborers worked 7 days per week rather than the 6 days per week that they worked during non-peak time periods. Workday and fieldwork information are used to determine feasible acreage farmed with machinery inventory. Labor resources for the 1,243 cropping acres include two individuals at permanent status who can work 12 hour days. There were no temporary or paid employees in the baseline model. Next, farm machinery were inventoried.    

Machinery inventory generally includes tractors, implements, and harvesters. Three tractors were available to operate up to 10 hours per day. Machinery operation requires one labor hour per machine hour with the exception of the sprayer and harvesting equipment (Table 1). The sprayer operation requires one person to operate the sprayer and one additional person to haul water. During harvest, labor was needed to operate the combine, grain cart, and trucks necessary to transport grain. Therefore, for each hour of combine operation, three labor hours are required for grain cart and hauling (Table 1). For all production activities, acres per hour for each machine were calculated, and the number of persons needed to operate each machine were assigned (i.e. for planting operations one person may be on the planter and one-half person assigned to filling seed for total of 1.5 people per hour). Machinery operations for land preparation, planting, post-planting, and harvesting activities were identified. Field operations with machinery are listed for the corn preceded by soybean phase in Table 2. Rosburg Farms produces continuous corn and a rotation of corn and soybean. Minimum corn acreage preceded by soybean was set at 621.5. A three-year rotation of corn preceded by soybean, corn preceded by corn, and soybean preceded by corn was set at minimum acres of 16.7 each.

 

Table 1: Machinery Resources of Rosburg Farms: Machinery and Labor Hours.

 Machinery 

No. of machines

Total hours per day

Tractor required

Tractor hours per machine per hour

Labor hrs. per machine per hour

 Chisel

2

10

Big Tractor

1

1

 Disc

1

10

Big Tractor

1

1

 Field Cultivator

1

10

Big Tractor

1

1

Planter

1

10

Big Tractor

1

1.5

 Sprayer

1

10

Small Tractor

1

1.2

 Cultivator

1

10

Small Tractor

1

1

 Combine

1

10

Big Tractor

1

3

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 2: Corn Preceded by Soybean Field Operations.

 Production phase

Machinery type

Beginning period

Ending period

Working rate: acres per hour

 Land Preparation

Field Cultivator

Dec. 6- April 21

May 3- May 9

30

 Planting

Planter

April 22-April 25

May 31- June 6

12

 Post-Plant

Sprayer

3 weeks after planting

1 week to complete

50

 Harvest

Combine

Sept 20- Sept. 26

Nov. 15- Dec. 5

12

 

 

 

 

 

 

 

 

 

 

Yield response to planting and harvesting dates were required for model parameterization. Expected yields when crops were planted and harvested in the best time period were assigned based on farm-level five-year average yield. For acres planted in the soybean preceded by corn rotation, the best yield was set at 55 bushels per acre. Expected yield for corn following soybean were assigned a per bushel yield of 200. Best yield for the continuous corn acreage was set at 180 bushels per acre. Continuous corn yield takes into consideration a yield lag expected for second-year corn which was a 10% reduction from the expected 200 bushel per acre yield.

Region-specific Constraints

Before models could be run, region-specific constraints must be assigned. A detailed description of constraints specific to this LP model can be found in Griffin and Lowenberg-DeBoer (2017). The most notable region-specific constraint is fieldwork probability. In particular, the probability of the number of days per week fieldwork can be conducted must be estimated given historic weather and soil conditions (Carls and Griffin, 2016). Rather than using median or average fieldwork days, days per week calculated at the 30th percentile for northwest Iowa were chosen (Doster et al., 2010) (Table 3). Detailed instructions on acquiring and analyzing USDA NASS data are presented in Griffin (2009) and Mensing (2017). In order to optimize yields, producers desire to conduct fieldwork in a timely manner while being efficient with machinery and labor resources (Griffin and Barnes, 2017). Farm management implications of days suitable for fieldwork (DSFW) include acreage that can be planted and/or harvested given available machinery resources.

 

Table 3: Days Suitable for Fieldwork in Northwest Iowa.

 Time period

Period length (weeks)

Good field days per period

Labor days per week

 December 6-April 21

19.5

9.3

7

 April 22- April 25

0.5

1.3

7

 April 26-May 2

1

3.1

7

 May 3-May 9

1

3.2

7

 May 10-May 16

1

3.3

7

 May 17-May 23

1

3.9

7

 May 24-May 30

1

3.8

7

 May 31-June 6

1

3.6

7

 June 7-June 13

1

4.3

7

 June 14-June 20

1

4.5

6

 June 21-June 27

1

4.4

6

 June 28-July 4

1

4.5

6

 July 5-July 11

1

4.7

6

 July 12-August 29

7

30.5

6

 August 30-September 19

3

14.6

6

 Sept. 20- Sept. 26

1

4.4

7

 September 27-October 10

2

9.3

7

 October 11- October 31

3

14.1

7

 November 1-November 14

2

8.1

6

 November 15- Dec. 5

3

9.9

6

                                                                                                                                              

Results

Linear Programming Model Results

Total contribution margin from the baseline LP model was $504,211, which does not include opportunity cost of land, unpaid labor, management, or machinery. Contribution margin can be considered returns to land, unpaid labor, and machinery. Planter machinery resources were limiting, although some capacity remains. Baseline farm operation was considered timely with crops being planted and harvested during acceptable time periods.

Additional crop acreage were added for the second model run holding all other constraints constant. The second run added 384 acres to the base farm of 1,243 for a total 1,618 acres. Results of modified models with additional acreage indicated land was able to be fully utilized. Comparison of baseline labor results to the modified model revealed full-time labor was utilized earlier in the spring and required more hours to conduct all field operations. Available hours for each machine were calculated based on field activities that occurred in the specified time period, labor availability, and days suitable for fieldwork.

The case study farm operation was considered timely with the baseline constraints of labor, machinery, crop rotation, and land for crop production. In the modified LP scenario with additional land for crop production, results indicated all land was planted and harvested with the same laborers and machinery resources defined in the baseline model. If labor or machinery ever decreased, timeliness, yield, and income from the crop production would be impacted. Results from the baseline and modified model runs were evaluated by financial analysis.

Crop Enterprise Budgeting Results

Enterprise budgets were developed reflective of 2017 growing season estimates. The annual enterprise budget utilized different values compared to the 5-year linear programming planning model. The Herbicide Tolerant Soybean following Corn Cost Production budget was generated on 613 acres with an anticipated yield of 55 bushels per acre and compared to the scenario of additional 801 crop acres allocated to soybean production. The planter expense included additional costs for variable-rate seeding. At Rosburg Farms, operators do one application of herbicide using the sprayer and a local cooperative custom applies the second application. Interest was calculated at 4.8% for 12 months on the operating note. The operating note was used to purchase inputs and repaid after crops were harvested. Harvesting costs were calculated based on custom rates including machinery, labor, and fuel cost for the combine, grain cart, and hauling, plus an additional cost for global navigation satellite system (GNSS) mapping. Total cost for modified acres represents the model scenario of additional 801 rented soybean acres.

Total cost per acre to produce soybean was $473.81. Cash rent equivalent of $225 was used based on average from 2016 Iowa State University Extension Cash Rent survey for the case study county (Plastina et al., 2016). Corn cost of production budget was generated on 613 acres with an anticipated yield of 200 bushels per acre. Results were compared to the scenario of 817 acres allocated to corn production from additional land rental.  When machinery costs were calculated, additional costs for variable-rate seeding were included for planting. Similar to soybean budgets, sprayers were used for one application of fertilizer and a local cooperative was hired to custom apply the second application. Interest was 4.8% for a 12-month operating note. Harvest costs were calculated based on complete custom rate per acre which included machinery, labor, and fuel cost for combine, grain cart and hauling to farm storage, plus an additional cost for GNSS yield mapping. Total cost for additional acres represents the model scenario of 817 corn acres. The cost per acre for corn production was $658.21, and calculated using custom machinery rates including labor, maintenance, and fuel (Table 4).

 

Table 4:  Production Cost Budget: Corn following Soybean.

 

Cost per acre

Total cost for 630 acres

Total cost 817 acres

Tillage and Planting Machinery

     

Field Cultivate

$14.05

$8,852

$11,479

Planter

$24.40

$15,372

$19,935

Spray

$6.00

$3,780

$4,902

Custom Application: Fertilizer

$6.00

$3,780

$4,902

Total Machinery Cost

$50.45

$31,783

$41,218

Seed, Chemicals, etc.

     

Seed

$106.42

$67,045

$86,945

Nitrogen

$60.00

$37,800

$49,020

Phosphate

$38.50

$24,255

$31,455

Potash

$21.00

$13,230

$17,157

Herbicide

$38.10

$24,003

$31,128

Crop Insurance

$12.20

$7,686

$9,967

Miscellaneous

$10.00

$6,300

$8,170

Interest

$16.16

$10,181

$13,203

Total input cost

$302.38

$190,499

$247,044

Harvest Machinery

     

Combine

$37.20

$23,436

$30,392

Drying price per bushel

$36.40

$22,932

$29,738

Handling grain by auger price per bushel

$6.78

$4,271

$5,539

Total Harvest Machinery Cost

$80.38

$50,639

$65,669

Land

     

Cash rent equivalent

$225

$137,970

$183,825

Total All Costs

$658.21

$410,891

$537,756

 

Net returns for corn production were calculated based on the fall 2017 price at the study area’s local cooperative’s price of $3.64 and 200 bushel per acre yield. This resulted in a $47,748 return at the baseline acres of 630 and a $57,019.49 return for additional rented acres for crop production (Table 5). Soybean net returns were calculated based on a 55 bushel per acre yield and a $9.80 per bushel price. Returns for soybean crop were $39,352.15 for the baseline 613 acres. In the modified scenario of acres planted for soybean production there was a return of $51,340.

 

 

Table  5: Projected Net Returns for Corn and Soybean Production.

 

Corn

Soybean

Baseline

   

Acres

630

613

Expected Selling Price/ Bushel

 $             3.64

 $                  9.80

Baseline Model

   

Total Returns

$458,640

 $       330,515.00

Total Costs

$410,891.90

 $       291,162.85

Baseline Model Net Returns

 $    47,748.10

 $         39,352.15

Acres

817

801

Modified Scenario

   

Total Returns

 $ 594,776.00

 $       431,739.00

Total Costs

$537,756.51

 $       380,398.45

Modified Scenario Net Returns

 $    57,019.49

 $         51,340.55

                      

 

Discussion

Given the feasibility of increasing farm acreage, financial analyses of additional acreage were conducted. Income for the farm comes from sales of corn and soybean produced the previous fall. Income was based on the projected net incomes calculated based on 200 bushels per acre yield on corn and 55 bushel per acre yield on soybean.  Machinery expenses were assigned based on custom rates. Rent payments vary between contracts and different landowners; however, an average $225 per acre was assumed. Storage costs for grain were calculated by taking storage capacity available per bushel minus amount of grain produced. This was the amount of grain needed to be sold in the fall or stored off farm. An average of $0.04 per bushel per month was used for the off-farm storage rate. One-third of storage expenses were paid up front in November and remainder was calculated and expensed in the months the crops were sold. Crop insurance was figured as an expense of $2.86 per acre for corn and $3.59 per acre for soybean. Rates were based on quotes from an area crop insurance agent. Simple breakeven analysis was conducted to determine corn price needed to justify farming full time. Based on breakeven analysis, corn would need to be $4.12 per bushel.

The farm could feasibly grow by 384 acres, but family living expenses need to be considered when determining if the farm could sustain the youngest family member as a full-time farmer. Based on a family size of four, estimated annual living expenses was $135,653 (Kraph et al., 2016). Farm income was based on the high third of families surveyed and includes medical expenses, insurance, and expendables. Family living expenses were also taken out for the older two decision makers. For a family of two, the estimated annual living expense $84,779 was assumed (Kraph et al, 2016). For the time being it was determined that off-farm employment were still necessary for the youngest family member, however the farm would be able to expand by 384 acres.

                                   

Conclusion

Analyses were conducted to ascertain the feasibility of adding a family member as a full-time farmer. Additional land rental was evaluated by a whole-farm planning model and intermediate results subjected to financial analysis. Results indicated Rosburg Farms could accommodate additional acres with existing machinery, however off-farm employment would still be necessary until the farm grew with respect to both acreage and equipment. If Rosburg Farms expand farm size in the future, increasing planting capacity may be necessary to ensure timely planting.

 

References

Boehlje, M. D., and Eidman, V.R. (1984). Farm Management. New York: Wiley.

Carls, E., and Griffin. T.W. (2016). Development of interactive website charts for utilization using Google Docs. Journal of Extension, 54(5).

Dantzig, G.B. (1949). Programming interdependent activities, II, mathematical model. Econometrica, 17(3/4):200-211.

Dobbins, C.L., Doster, D.H., Patrick, G.F., Miller, W.A., and Preckel, P.V. (2006). Purdue PC-LP Farm Plan B-21 Crop Input Form. Purdue University.

Doster, D. H., Dobbins, C. L., Griffin, T. W., & Erickson,B. (2010). B-21 input form guide book. C-EC-11-Rev. Department of Agricultural Economics, Purdue University.

Griffin, T.W., and Barnes, E. (2017). Available time to plant and harvest cotton across the cotton belt. Journal of Cotton Science, 21(1):8-17.

Griffin, T. W., Mark, T. B., Dobbins, C. L., and Lowenberg-Deboer, J. M. (2014). Estimating whole farm costs of conducting on-farm research on Midwestern US corn and soybean farms: A linear programming approach. International Journal of Agricultural Management, 4(1):21-27.

Griffin, T.W. (2009). Acquiring and applying days suitable for fieldwork for your state. Journal of the American Society of Farm Managers and Rural Appraisers, 72(1):35-42

Griffin, T.W., Dobbins, C.L., Florax, R.J.G.M., Lowenberg-DeBoer, J.M., and Vyn, T.J. (2010). Spatial analysis of precision agriculture data: Role for Extension. Journal of the National Association of County Agricultural Agents, 3(1).

Griffin, T.W. and Lowenberg-DeBoer, J.M. (2017). Impact of automated guidance for mechanical control of herbicide resistant weeds in corn. Journal of Applied Farm Economics, 1(2):62-74.

Johanns, A. (2017). Ag decision maker crop decision tools. Iowa State University Extension and Outreach. Iowa State University. Accessed February 4, 2017. https://www.extension.iastate.edu/agdm/decisionaidscd.html.

Kraph, B. M., Raab, D., and Zwilling, B. L. (2016). Farm and family living income and expenditures. FarmDoc. Department of Agriculture and Consumer Economics. October 2016. Accessed February 19, 2017. http://www.farmdoc.illinois.edu/manage/enterprise_cost/FBM-0190familyliving.pdf.

Mensing, M. (2017). Farm management implications of uncertainty in days suitable for fieldwork. Master of Agricultural Business Thesis, Kansas State University. Accessed February 2017.

Plastina, A. (2017). Estimated cost of crop production in Iowa. Department of Economics, Iowa State University. Accessed February 11, 2017. https://www.extension.iastate.edu/agdm/crops/html/a1-20.html.

Plastina, A., Johnanns, A., and Welter, C. (2016). Cash rental rates for Iowa 2016. Survey, Department of Economics, Iowa State University. Accessed February 11, 2017. http://www.extension.iastate.edu/agdm/wholefarm/pdf/c2-10.pdf.

Rosburg, B. (2017). Inter-generational transition strategy assessment: The case of Rosburg farms. Master of Agricultural Business Thesis, Kansas State University. Available at http:// krex.k-state.edu/dspace/handle/2097/35381.

Tice, T. F. (1973). Farm machinery selection: A linear programming model. MS Thesis, Kansas State University. Accessed 2017.

Williams, J., and R. Llewelyn. (2013). Days suitable for field work in Kansas by crop reporting regions. Department of Agricultural Economics, Kansas State Research and Extension. Accessed 2016. https://www.agmanager.info/sites/default/files/FieldWorkdays_Kansas.pdf.

 

Acknowledgements

The authors appreciate constructive criticism from three anonymous reviewers and the editor. Invaluable informal reviews were provided by Emily Carls, Jared Cullup, Karli Pryor, and Dana Griffin. The USDA NASS provided primary data including days suitable for fieldwork. We are thankful to the Rosburg family members, Brian, Richard, and Keith for their contribution to this original research and thesis. The thesis was in partial fulfillment of the requirement of the Masters of Agribusiness (MAB) program in the Department of Agricultural Economics at Kansas State University, for whom we are grateful.