Teaching Chapter 18: Product Development Economics
Timing
This session can be placed almost anywhere in the sequence of a product development course. For a project course, since there are many essential topics necessary to cover for concept development, this session can easily be placed after the concept development material is completed. We generally place this session about two thirds of the way through our course.
Objectives and Strategy
This session can be used to prepare students to handle the inevitable economic decisions facing development projects. Such decisions include go/no-go decisions at various points in the development effort and sensitivity analyses conducted to evaluate many types of trade-offs. Management students will typically be familiar with NPV analysis and the use of spreadsheets to facilitate this. On the other hand, this type of analysis may be new to most engineering students and industrial design students. In an integrated course, careful thought must be given to how to engage the entire class.
Session Outline
The session can follow this flow:
Introduction to Financial Analysis of Projects
We like to begin the class by reviewing the types of decisions that can be supported using financial analysis of a project:
It is then useful to review the basic concept of NPV and how this can be used to construct financial analysis models of projects. The NPV formula can be written on the board. This method provides a means for computing economic payback and for combining the projected cash flow with the time value of the cash. One can then ask the class what alternative economic models or metrics might be useful instead. Students will usually respond with other financial metrics such as return on investment (ROI). They should also be encouraged to consider time-based metrics such as payback period, break-even time, first profitable shipment, etc. A graph similar to Exhibit 15-2 can be drawn to illustrate these time-based metrics, as shown below.
This session can be placed almost anywhere in the sequence of a product development course. For a project course, since there are many essential topics necessary to cover for concept development, this session can easily be placed after the concept development material is completed. We generally place this session about two thirds of the way through our course.
Objectives and Strategy
This session can be used to prepare students to handle the inevitable economic decisions facing development projects. Such decisions include go/no-go decisions at various points in the development effort and sensitivity analyses conducted to evaluate many types of trade-offs. Management students will typically be familiar with NPV analysis and the use of spreadsheets to facilitate this. On the other hand, this type of analysis may be new to most engineering students and industrial design students. In an integrated course, careful thought must be given to how to engage the entire class.
Session Outline
The session can follow this flow:
- Introduction to Financial Analysis of Projects
- Review of Financial Analysis Methodology
- Principles and Caveats
- In-Class Exercise
- Discussion
Introduction to Financial Analysis of Projects
We like to begin the class by reviewing the types of decisions that can be supported using financial analysis of a project:
- Go/No-Go Decisions
- e.g., Should we proceed with the project?
- Sensitivity Analysis
- e.g., What if the base-case assumptions are incorrect?
- Trade-off Decisions
- e.g., Should we spend more time on DFM to reduce production cost?
- Qualitative Factors
- e.g., How will the project be affected by changes in the competitive environment?
It is then useful to review the basic concept of NPV and how this can be used to construct financial analysis models of projects. The NPV formula can be written on the board. This method provides a means for computing economic payback and for combining the projected cash flow with the time value of the cash. One can then ask the class what alternative economic models or metrics might be useful instead. Students will usually respond with other financial metrics such as return on investment (ROI). They should also be encouraged to consider time-based metrics such as payback period, break-even time, first profitable shipment, etc. A graph similar to Exhibit 15-2 can be drawn to illustrate these time-based metrics, as shown below.
Adapted from Charles H. House and Raymond L. Price, "The Return Map: Tracking Product Teams", Harvard Business Review, January-February 1991, pp. 92-100.)
We then like to review the financial analysis method by discussing the principles and some caveats relating to each of the four steps described in the chapter.
Step 1: Build a Base-Case Model
We start by reviewing the inputs required for the financial model. These inputs are listed in Exhibit 17-3 and their timing is displayed in Exhibit 17-4.
The base-case NPV is computed using the NPV formula. It is useful for the class to discuss where the discount rate comes from. Some financial analysts will argue that the discount rate should be the risk-adjusted cost of capital or the rate of return for other projects of equivalent risk. Nevertheless, most firms use a single cost of capital for all projects. This discussion brings up the issue of project risk (a complex issue involving technical, marketing, and many other concerns) and its relationship to financial concerns.
At this early stage it is useful to begin to "sanity check" the process by asking one or more of the following questions: (Note that the third of these questions is the basis of Exercise 17.1.)
The answers to these questions depend upon several factors that students should list, including: risk, uncertainty, sensitivity, investment size, coupling with other projects or products, learning value of the project, and other qualitative factors.
Exercise 17.2 can be discussed at this point in the session. (See the sample solution.) Alternatively, the construction of a financial model can be left for the in-class exercise later.
Step 2: Sensitivity Analysis
Sensitivity analysis addresses the "what-if" questions, such as:
The sensitivity analysis procedure is simple. First the base-case NPV is saved. Then the model is modified to represent alternative scenarios. The results are displayed in tables as shown in Exhibits 17-9, 17-11, and 17-13.
Discussion for this step of the method can begin with the observation that some of the NPV sensitivities are linear (e.g., development cost, Exhibit 17-9), while other sensitivities are nonlinear (e.g., development time, Exhibit 17-11). The answer is due to the exponential nature of the compounding of interest. Thus shifting cash flows in time has a nonlinear effect on NPV, whereas changing any of the cash flow numbers themselves has linear effects.
Step 3: Trade-off Analysis
There are six generic types of trade-offs to be considered, as illustrated by Exhibit 17-12. It may be useful to mention a scenario exemplifying each type of trade-off. While such examples do not always hold true, some possibilities are listed below:
Step 1: Build a Base-Case Model
We start by reviewing the inputs required for the financial model. These inputs are listed in Exhibit 17-3 and their timing is displayed in Exhibit 17-4.
- development cost
- ramp-up cost
- marketing and support cost
- production cost (unit variable cost)
- production volume (usually equal to the sales volume)
- revenue (unit price)
- timing and schedule of above cash flows
The base-case NPV is computed using the NPV formula. It is useful for the class to discuss where the discount rate comes from. Some financial analysts will argue that the discount rate should be the risk-adjusted cost of capital or the rate of return for other projects of equivalent risk. Nevertheless, most firms use a single cost of capital for all projects. This discussion brings up the issue of project risk (a complex issue involving technical, marketing, and many other concerns) and its relationship to financial concerns.
At this early stage it is useful to begin to "sanity check" the process by asking one or more of the following questions: (Note that the third of these questions is the basis of Exercise 17.1.)
- Do we always "go" if NPV is positive?
- Do we ever "not go" if NPV is positive?
- Do we ever "go" if NPV is negative?
- Do we always "not go" if NPV is negative?
The answers to these questions depend upon several factors that students should list, including: risk, uncertainty, sensitivity, investment size, coupling with other projects or products, learning value of the project, and other qualitative factors.
Exercise 17.2 can be discussed at this point in the session. (See the sample solution.) Alternatively, the construction of a financial model can be left for the in-class exercise later.
Step 2: Sensitivity Analysis
Sensitivity analysis addresses the "what-if" questions, such as:
- What if we spend 10% more on development?
- What if we reduce the price after Year 2?
- What if we take an extra four months in development?
- What if we sell half as many as planned?
The sensitivity analysis procedure is simple. First the base-case NPV is saved. Then the model is modified to represent alternative scenarios. The results are displayed in tables as shown in Exhibits 17-9, 17-11, and 17-13.
Discussion for this step of the method can begin with the observation that some of the NPV sensitivities are linear (e.g., development cost, Exhibit 17-9), while other sensitivities are nonlinear (e.g., development time, Exhibit 17-11). The answer is due to the exponential nature of the compounding of interest. Thus shifting cash flows in time has a nonlinear effect on NPV, whereas changing any of the cash flow numbers themselves has linear effects.
Step 3: Trade-off Analysis
There are six generic types of trade-offs to be considered, as illustrated by Exhibit 17-12. It may be useful to mention a scenario exemplifying each type of trade-off. While such examples do not always hold true, some possibilities are listed below:
Positive Effect
|
Negative Effect
|
Example Scenario
|
development time down
|
development cost up
|
hiring more engineers to "crash" the project
|
product cost down
|
development cost up
|
conducting more DFM iterations
|
product quality up
|
product cost up
|
adding a costly fortune
|
product quality up
|
development time up
|
taking more time for testing and optimization
|
product cost down
|
development time up
|
conducting more DFM iterations
|
development cost down
|
product quality down
|
"cutting some corners" to save money
|
Computing trade-offs is also quite straightforward. First compute the NPV gain of the positive effect, then compute the NPV loss of the negative effect. Finally, compare the two effects to decide if the trade-off is worthwhile.
Trade-off rules can inform many decisions made by members of the project team on a daily basis. For example, knowing that "a $1 increase in product cost means a $43k loss in NPV" can put some DFM decisions into perspective (from Exhibit 17-14). These rules are simply the sensitivities computed in step 2.
Exercise 17.3 can be discussed at this point in the session, or the computation of trade-offs can be left for the in-class exercise later in the session.
A discussion of the effects of delaying product launch is worthwhile at this point. Thought Question 17.2 proposes two possible models for the impact of delaying a product introduction. Sales can either be shifted later in time or lost forever. Both of these scenarios can easily be modeled, and there are other possible models to consider as well. (See the sample solution.)
Step 4: Qualitative Factors
It is important for students to realize that the quantitative analysis presented above focuses on measurable financial quantities only. To facilitate numerical analysis, the models must make many assumptions. Such analysis ignores intangibles like brand equity, customer loyalty, and organizational learning. Just because the spreadsheets are capable of computing precise results does not guarantee that the results are accurate.
The term "qualitative factors" may not be the most appropriate term for the miscellaneous factors we consider in this step. Many of these effects may in fact be quantifiable. For example, a financial analysis can certainly be done for an entire product line rather than at the level of the individual product. Sensitivities can be computed for exchange rates, material prices, etc. Still other effects will remain unquantifiable, such as social and economic trends. Nevertheless, if one can anticipate which factors may possibly be important, various "what-if" scenarios can be considered to understand the possible impact on the financial results.
The topics of Thought Questions 17.1 and 17.3 can be discussed at this point in the session.
Props
Almost any products can be discussed as examples in class. It is best to bring a product where the relevant financial data can be obtained or realistically guessed.
In-Class Exercise
We have found it useful to illustrate the use of spreadsheet-based financial analysis in class. Such an exercise can be based on any one of the following:
The exercise requires the use of a computer connected to a projection display enabling the class to see the screen display. It is helpful to set up the spreadsheet ahead of time. Then the class can simply fill in the blanks and see the results. If the class is relatively unfamiliar with spreadsheet usage, it can be helpful to start from a new worksheet and to actually input the formulas live rather than setting them up in advance.
The Polaroid example is detailed in the text. The bicycle light example is worked out in the sample solution for Exercises 17.2 and 17.3.
Another possible example spreadsheet is shown below for a student project. This spreadsheet layout allows all of the inputs to be entered in the lower table and the cash flow table is automatically generated. Then once the base case NPV is saved, deviations can be entered to compute sensitivities, trade-offs, and more complex scenarios. This spreadsheet is available to interested faculty to use in class and may be obtained via the web site.
Trade-off rules can inform many decisions made by members of the project team on a daily basis. For example, knowing that "a $1 increase in product cost means a $43k loss in NPV" can put some DFM decisions into perspective (from Exhibit 17-14). These rules are simply the sensitivities computed in step 2.
Exercise 17.3 can be discussed at this point in the session, or the computation of trade-offs can be left for the in-class exercise later in the session.
A discussion of the effects of delaying product launch is worthwhile at this point. Thought Question 17.2 proposes two possible models for the impact of delaying a product introduction. Sales can either be shifted later in time or lost forever. Both of these scenarios can easily be modeled, and there are other possible models to consider as well. (See the sample solution.)
Step 4: Qualitative Factors
It is important for students to realize that the quantitative analysis presented above focuses on measurable financial quantities only. To facilitate numerical analysis, the models must make many assumptions. Such analysis ignores intangibles like brand equity, customer loyalty, and organizational learning. Just because the spreadsheets are capable of computing precise results does not guarantee that the results are accurate.
The term "qualitative factors" may not be the most appropriate term for the miscellaneous factors we consider in this step. Many of these effects may in fact be quantifiable. For example, a financial analysis can certainly be done for an entire product line rather than at the level of the individual product. Sensitivities can be computed for exchange rates, material prices, etc. Still other effects will remain unquantifiable, such as social and economic trends. Nevertheless, if one can anticipate which factors may possibly be important, various "what-if" scenarios can be considered to understand the possible impact on the financial results.
The topics of Thought Questions 17.1 and 17.3 can be discussed at this point in the session.
Props
Almost any products can be discussed as examples in class. It is best to bring a product where the relevant financial data can be obtained or realistically guessed.
In-Class Exercise
We have found it useful to illustrate the use of spreadsheet-based financial analysis in class. Such an exercise can be based on any one of the following:
- The Polaroid Digital PhotoPrinter example given in the text
- The hypothetical bicycle light scenario given in Exercises 17.2 and 17.3
- Another product example to fill in during the class (We suggest a popular product that the class would be familiar with, such as any of the products listed in Chapter 1 Exhibit 1-2.)
The exercise requires the use of a computer connected to a projection display enabling the class to see the screen display. It is helpful to set up the spreadsheet ahead of time. Then the class can simply fill in the blanks and see the results. If the class is relatively unfamiliar with spreadsheet usage, it can be helpful to start from a new worksheet and to actually input the formulas live rather than setting them up in advance.
The Polaroid example is detailed in the text. The bicycle light example is worked out in the sample solution for Exercises 17.2 and 17.3.
Another possible example spreadsheet is shown below for a student project. This spreadsheet layout allows all of the inputs to be entered in the lower table and the cash flow table is automatically generated. Then once the base case NPV is saved, deviations can be entered to compute sensitivities, trade-offs, and more complex scenarios. This spreadsheet is available to interested faculty to use in class and may be obtained via the web site.