ݺߣ

ݺߣShare a Scribd company logo

Here’s how to decide between stumpage- or delivered price harvest scheduling models

•
•
KR Walters Consulting Services
KR Walters Consulting Services

In this article, I will explore these the stumpage-price and delivered-priced harvest scheduling models, first by delving into the mechanics involved in building them, and then by discussing their implications for managers. Of the two, a stumpage-price model is the more common and easier model to formulate.

1 of 2
Download to read offline
From the March 2007 issue of The Forestry Source Here’s How to... Decide between Stumpage- or Delivered-Price Harvest Models By Karl Walters, CF are applied in the same fashion as in the stumpage model. In the stumpage model, the destination for products is predetermined; in W hen I meet with a client to discuss the details of a har- the delivered- price model, the destination for products is part of vest scheduling modeling project one of the first ques- the decision. tions I ask is, “What are your objectives?” Most often, For example, in a stumpage model, a decision variable may the answer is to maximize the value of discounted future cash represent a clearcut harvest in period 1, yielding 100 flows. In these cases, the next question is, “Do you want a MBF of timber, and revenue of $147.50. In the delivered-price stumpage- or a delivered-price model?” While the answer clearly model, there may be multiple decision variables, all with the same has implications for the model-building process, there is a more 100 MBF yield, harvest cost, and severance tax coefficients, but fundamental question being asked about the client’s management with different hauling cost coefficients. Because the revenue coef- goals tha, he or she may not have considered. ficient in the stumpage model is an average, the net revenue gen- In this article, I will explore these two models, first by delving erated in the delivered price model for that same stand may be into the mechanics involved in building them, and then by dis- quite different. cussing their implications for managers. Of the two, a stumpage- Delivered-price models can get very complicated, depending price model is the more common and easier model to formulate, on the number of products, mill destinations, and how detailed one so I will start there. gets calculating hauling costs (e.g. stand-level vs. tract-level). In a stumpage-price model, the prices for various products rep- Remsoft’s Allocation Optimizer extension to Woodstock is resent the average value of timber and the total of harvesting and designed specifically to address complex delivered-price models. hauling costs, as well as other costs, such as severance taxes levied So, let’s go back and look at the decision variables again, but at the stump. For example, suppose a nearby mill pays $350 per this time I will examine what it means from a managerial perspec- MBF for conifer saw logs delivered to the gate. If it costs $120 to tive to prefer a stumpage model over delivered price. harvest 1 MBF, $80 to haul 1 MBF to the mill, and the state impos- In a stumpage model, management is asking, “What is the best es a severance tax of $2.50 per MBF, the net return to the landown- strategy to manage this property to maximize financial returns?” If er is $147.50 per MBF. two stands have the same species, site, and stocking, the silvicul- If there are multiple mills in the area with various hauling dis- tural regime to be applied to them is exactly the same, regardless tances, a weighted average stumpage price can be determined for of their location in the forest. The viewpoint is distinctly forest- each product. There also may be different stumpage prices, level, and is typical of large landholders such as integrated forest depending on whether the timber is produced via thinning or final products companies that historically owned forestland to meet harvest. And, if the forest estate being modeled is large and geo- their own mill requirements. graphically diverse, stumpage prices may be assigned at the tract In a delivered-price model, the management question is level, representing logical assignments of logs to the nearest mill focused more at the stand or tract level: “What is the best invest- to minimize hauling costs. ment to apply to this stand or tract to maximize financial returns While stumpage represents the revenue side of the equation, overall?” Consider the same two stands in the previous example, silvicultural treatments provide the costs: site preparation, plant- except that one stand is 10 miles from the nearest mill and the ing, herbaceous competition control, fertilization, precommercial other is 80 miles. Does it make sense to invest as much silvicul- thinning, pruning, etc. The decision variables in a stumpage model ture in the remote stand as the nearby stand? The optimal solution determine whether an activity (e.g., fertilization, commercial thin, probably suggests no. final harvest) is carried out, and when and where it is implement- So other than for the sake of simplicity, why would anyone ed on the forest. These decision variables trigger revenue or cost choose a stumpage model over a delivered price model? Consider outputs. In the objective function, the discounted silvicultural that harvest scheduling models are strategic planning tools, with costs are subtracted from the discounted harvest revenues to yield long planning horizons, and that the most consistent aspect of any the net present value of future cash flows. planning environment is change. Suppose that next year, the near- In a delivered-price model, the prices for various products rep- by mill from the previous example closes, and now the two stands resent what a particular mill pays at the gate (e.g., $350 per MBF are equally distant from another mill. Would your silvicultural from our previous example). If there are multiple mills to consid- strategy change? With the stumpage model, changes to the silvi- er, each may have a unique price for each product it buys. cultural strategy are not likely to change much at all, but a com- Moreover, the hauling distance to each mill will similarly be pletely different strategy would probably result in the delivered- unique and therefore will have a unique, associated hauling cost. price model. At the stand level, harvesting and silvicultural costs are not affect- Moreover, foresters charged with implementing silviculture ed by the decision to send products to any one mill. These costs typically do not maintain different sets of silvicultural regimes for
each forest tract. Instead, they are more likely to use a common set of prescriptions over the entire land base to maximize productivi- ty overall, with the hope of creating more options for selling into a changeable marketplace. After all, a chip-n-saw log can be sold as pulpwood if need be, but the converse is not true. So which model form do you choose? It depends on several factors, but land tenure is a key consideration. If you represent a timberland investment management organization and the forest is being held in a closed-end fund for 10 years, your planning hori- zon is sufficiently short so that events like a mill closure would force large-scale changes to your strategy anyway, and a deliv- ered-price model that is sensitive to these types of changes may be appropriate. On the other hand, if you have a long- term outlook and plan to hold land for decades, there is the very real possibility that a new mill could open to replace the one that closed. If the forest has been managed to consistent standards throughout, it is less likely that you will have made a very wrong decision about a particular stand (e.g., high site stand under minimum-level management five miles from the new mill), than with a delivered-price model. Also of importance are the number and diversity of potential mills to which wood could be delivered, both within your organization and to outside buyers. Increases to the number of mills available increase the likelihood that there is a mill close to any stand or tract, which makes the deci- sions similar to a stumpage-price model. The final decision is a trade-off between precision at the tactical level versus a more strategic look over the long term. A final decision can only be reached by clearly understanding the goals and objectives of the planning exercise and with a thorough knowledge of the forest and the markets for wood products in the region. Walters is vice-president of forest planning services for FORSight Resources. For more information, contact him at FORSight Resources, LLC3813 H Street, Vancouver, WA 98663; (360) 882-9030; karl.walters@forsightresources.com; http://for- sightresources.com.

More Related Content

Here’s how to decide between stumpage- or delivered price harvest scheduling models

  • 1. From the March 2007 issue of The Forestry Source Here’s How to... Decide between Stumpage- or Delivered-Price Harvest Models By Karl Walters, CF are applied in the same fashion as in the stumpage model. In the stumpage model, the destination for products is predetermined; in W hen I meet with a client to discuss the details of a har- the delivered- price model, the destination for products is part of vest scheduling modeling project one of the first ques- the decision. tions I ask is, “What are your objectives?” Most often, For example, in a stumpage model, a decision variable may the answer is to maximize the value of discounted future cash represent a clearcut harvest in period 1, yielding 100 flows. In these cases, the next question is, “Do you want a MBF of timber, and revenue of $147.50. In the delivered-price stumpage- or a delivered-price model?” While the answer clearly model, there may be multiple decision variables, all with the same has implications for the model-building process, there is a more 100 MBF yield, harvest cost, and severance tax coefficients, but fundamental question being asked about the client’s management with different hauling cost coefficients. Because the revenue coef- goals tha, he or she may not have considered. ficient in the stumpage model is an average, the net revenue gen- In this article, I will explore these two models, first by delving erated in the delivered price model for that same stand may be into the mechanics involved in building them, and then by dis- quite different. cussing their implications for managers. Of the two, a stumpage- Delivered-price models can get very complicated, depending price model is the more common and easier model to formulate, on the number of products, mill destinations, and how detailed one so I will start there. gets calculating hauling costs (e.g. stand-level vs. tract-level). In a stumpage-price model, the prices for various products rep- Remsoft’s Allocation Optimizer extension to Woodstock is resent the average value of timber and the total of harvesting and designed specifically to address complex delivered-price models. hauling costs, as well as other costs, such as severance taxes levied So, let’s go back and look at the decision variables again, but at the stump. For example, suppose a nearby mill pays $350 per this time I will examine what it means from a managerial perspec- MBF for conifer saw logs delivered to the gate. If it costs $120 to tive to prefer a stumpage model over delivered price. harvest 1 MBF, $80 to haul 1 MBF to the mill, and the state impos- In a stumpage model, management is asking, “What is the best es a severance tax of $2.50 per MBF, the net return to the landown- strategy to manage this property to maximize financial returns?” If er is $147.50 per MBF. two stands have the same species, site, and stocking, the silvicul- If there are multiple mills in the area with various hauling dis- tural regime to be applied to them is exactly the same, regardless tances, a weighted average stumpage price can be determined for of their location in the forest. The viewpoint is distinctly forest- each product. There also may be different stumpage prices, level, and is typical of large landholders such as integrated forest depending on whether the timber is produced via thinning or final products companies that historically owned forestland to meet harvest. And, if the forest estate being modeled is large and geo- their own mill requirements. graphically diverse, stumpage prices may be assigned at the tract In a delivered-price model, the management question is level, representing logical assignments of logs to the nearest mill focused more at the stand or tract level: “What is the best invest- to minimize hauling costs. ment to apply to this stand or tract to maximize financial returns While stumpage represents the revenue side of the equation, overall?” Consider the same two stands in the previous example, silvicultural treatments provide the costs: site preparation, plant- except that one stand is 10 miles from the nearest mill and the ing, herbaceous competition control, fertilization, precommercial other is 80 miles. Does it make sense to invest as much silvicul- thinning, pruning, etc. The decision variables in a stumpage model ture in the remote stand as the nearby stand? The optimal solution determine whether an activity (e.g., fertilization, commercial thin, probably suggests no. final harvest) is carried out, and when and where it is implement- So other than for the sake of simplicity, why would anyone ed on the forest. These decision variables trigger revenue or cost choose a stumpage model over a delivered price model? Consider outputs. In the objective function, the discounted silvicultural that harvest scheduling models are strategic planning tools, with costs are subtracted from the discounted harvest revenues to yield long planning horizons, and that the most consistent aspect of any the net present value of future cash flows. planning environment is change. Suppose that next year, the near- In a delivered-price model, the prices for various products rep- by mill from the previous example closes, and now the two stands resent what a particular mill pays at the gate (e.g., $350 per MBF are equally distant from another mill. Would your silvicultural from our previous example). If there are multiple mills to consid- strategy change? With the stumpage model, changes to the silvi- er, each may have a unique price for each product it buys. cultural strategy are not likely to change much at all, but a com- Moreover, the hauling distance to each mill will similarly be pletely different strategy would probably result in the delivered- unique and therefore will have a unique, associated hauling cost. price model. At the stand level, harvesting and silvicultural costs are not affect- Moreover, foresters charged with implementing silviculture ed by the decision to send products to any one mill. These costs typically do not maintain different sets of silvicultural regimes for
  • 2. each forest tract. Instead, they are more likely to use a common set of prescriptions over the entire land base to maximize productivi- ty overall, with the hope of creating more options for selling into a changeable marketplace. After all, a chip-n-saw log can be sold as pulpwood if need be, but the converse is not true. So which model form do you choose? It depends on several factors, but land tenure is a key consideration. If you represent a timberland investment management organization and the forest is being held in a closed-end fund for 10 years, your planning hori- zon is sufficiently short so that events like a mill closure would force large-scale changes to your strategy anyway, and a deliv- ered-price model that is sensitive to these types of changes may be appropriate. On the other hand, if you have a long- term outlook and plan to hold land for decades, there is the very real possibility that a new mill could open to replace the one that closed. If the forest has been managed to consistent standards throughout, it is less likely that you will have made a very wrong decision about a particular stand (e.g., high site stand under minimum-level management five miles from the new mill), than with a delivered-price model. Also of importance are the number and diversity of potential mills to which wood could be delivered, both within your organization and to outside buyers. Increases to the number of mills available increase the likelihood that there is a mill close to any stand or tract, which makes the deci- sions similar to a stumpage-price model. The final decision is a trade-off between precision at the tactical level versus a more strategic look over the long term. A final decision can only be reached by clearly understanding the goals and objectives of the planning exercise and with a thorough knowledge of the forest and the markets for wood products in the region. Walters is vice-president of forest planning services for FORSight Resources. For more information, contact him at FORSight Resources, LLC3813 H Street, Vancouver, WA 98663; (360) 882-9030; karl.walters@forsightresources.com; http://for- sightresources.com.