Management of R&D - Reaction Paper No. 3

1. Financial Evaluation of R&D Projects

A. “Economic Analysis of R&D Projects: Real Option Versus NPV Valuation Revisited”

(Summary) There are two R&D project valuation models: 1) the traditional net present value (NPV) approach and 2) the real option approach. Recently, the option approach has been said to more accurately reflect, than NPV, the uncertainly and the sequential nature of the decision making situation for R&D managers. It is argued that an R&D project is like an investment call option because the firm has an option to wait for more information before deciding to commercialize (therefore, only the initial investment is at risk). However, using the option model means accepting assumptions from stock and derivatives pricing that may not necessarily apply to R&D projects. On the other, the proposed “NPV model is able to account for the convexity in the value of the project without bringing in the complex issues associated with evaluating stock price volatility of a “twin” security.” While both models take into account “the sequential nature of decisions involved in an R&D project”, the option approach requires a market proxy, whose volatility will be used to estimate a number of variables, to value a project. In most cases, this means finding an equivalent market-traded security that would behave like the R&D project “had it been traded on the market”. It is hard to be certain that this condition is fulfilled. The NPV approach does not need a market proxy, which makes it simpler and easier to understand. However, it does require direct estimates of certain variables. The estimates are expected to have significant subjectivity but the estimation process may prove a valuable exercise since it can help uncover risks and uncertainties in the project that may not be apparent. NPV considers the following input parameters: 1) initial investment, 2) cost to commercialize (at some future point), 3) opportunity cost, 4) probability of project success, 5) probability of a favorable market, 6) present values of future cash flow from commercialization for both favorable and unfavorable market and 7) option to commercialize or not.

(Reaction) Quantitative valuation models are useful tools but are of little value if used mechanically and out of context. Without understanding the theory or the implicit assumptions behind them, figures produced by these methods will give little insight into the projects they are supposed to evaluate and can even be misleading. Like any other method or model, it is imperative to evaluate the applicability or suitability of a quantitative valuation model before actually using it. To do this, the evaluators must understand both the nature of the product and the nature of the model.

B. “Real Options for Evaluating Venture Capital and Strategic R&D Investments”

(Summary) Integration of R&D and technology strategy into the overall corporate and business strategy requires the valuation of R&D investments. Due to their use of discount rates, traditional metrics like Net Present Value (NPV) make long-range, high uncertainty R&D look unattractive, which is contrary to what managers intuitively know (that high uncertainly means high future potential payoff). On the other hand, uncertainty is an integral component of the Real Options (RO) approach and is therefore, naturally accounted for in it. RO is founded on the option-pricing theory (a mathematical framework for valuing financial option contracts) and extends it to real or non-financial investments. Like a financial option, a real option grants the right, but not the obligation, to take an action in the future. While a financial option is detailed in a contract, a real option is embedded in the strategic investment (i.e. the firm has the right to act on and exploit opportunities its investment may produce in the future). Due to its theoretical foundation, RO allows “for a better melding of strategic intuition and analytical rigor.” Furthermore, the use of RO promotes a shift in outlook “from 'fear uncertainty and minimize investment' to 'seek gains from uncertainty and maximize learning' [that] can open up a wide range of possible—and profitable—actions. “ In general, RO is especially suitable when: 1) “uncertainty drives value”, 2) “there are a dynamic series of future decision points”, 3) there is “flexibility to adapt to changing business variables”, and 4) “management has credible mandate to respond to them and reallocate resources.” The value of RO comes not only from the numbers produced but from the thinking process and paradigm shift it calls for.

(Reaction) It is not clear that R&D projects behave like financial options on a fundamental level. There are certainly parallels but there are also significant differences. For one, R&D projects have on-going (or future, if the project is yet to be started) costs while financial options do not (i.e. cost is one-time and upfront during purchase of the option). Furthermore, the value of a financial option depends on the value of an underlying security. In fact, options are often used as hedge to limit potential losses due to exposure to the underlying security. This has no parallel in R&D projects. There is no underlying value upon which the value of an R&D project depends (in fact, the R&D project seeks to create this value). Finally, when an option is exercised, the owner of the option is guaranteed to get a value out of it (whether through a significant reduction in loss when used as an insurance or profit when use in itself). In other words, there is no uncertainty after the decision has been made to exercise (the uncertainty is on the occurrence of the conditions that would lead the owner to exercise his option). On the other hand, what does it mean to “exercise” an R&D project? There is uncertainty before and after the decision to commercialize. In other words, unlike exercising a financial option, one still does not exactly know how much he will get upon commercializing the product of an R&D project.

2. Selecting R&D Projects

A. “Research Decisionmaking in Industry: The Limits to Quantitative Methods”

(Summary) Review of quantitative analysis in R&D decision-making by individual firms showed that only very little systematic information is available and in most cases, the available information is anecdotal. Literature demonstrates the limited practical utility of quantitative methods and the industry's “reliance on subjective judgment and good communication between R&D, management and marketing staff in the decision-making process”. Like other investment opportunities, R&D competes for corporate support. R&D directors must be able to justify costs and demonstrate the value of projects to top management. In view of this, quantitative methods have been developed to review and evaluate “ongoing research, new research projects selection, and research resource allocation.” Few firms used quantitative methods to review ongoing research pointing out that they have tried but failed to come up with a good way to do it. Some even claim that it does more harm than good. Common macroeconomic measures like ROI, discounted cash flow and present value analysis have been found to bias against (and therefore unsuitable to review) long-term, high-risk projects like basic research. Furthermore, they favor short-term, high-yield projects which lead some scholar to argue that their use caused U.S. firms to under-invest in new technologies. Business opportunity techniques, which relies on future sales estimates and product development time, are also susceptible to substantial error and are also not useful in evaluating basic research (since it is difficult to assess its economic impact). Therefore, noted authors have argued that “rather than relying solely on quantitative techniques for the evaluation of R&D investments, managers must develop an understanding of the underlying technologies, and apply informed judgment in making such decisions.” In R&D project selection, quantitative methods fall under four main categories (in increasing sophistication in terms on input data and processing required): 1) scoring models, 2) economic models, 3) constrained optimization or portfolio models, and, 4) risk analysis or decision analysis models. Similar to evaluating on-going research, quantitative methods for evaluating and selecting new projects are not widely used. A primary deficiency that has been pointed out against quantitative methods is their high sensitivity to error (which is likely for high uncertainty projects making the methods' projections highly inaccurate). Furthermore, it has been observed that increased sophistication did not increase utility as they introduce as many limitations as they address. A study on resource allocation and strategic planning in general also observed that due to the complexity and diversity of the processes and hierarchies involved, there is no firm base to build and utilize a financial model. Rather, the study stressed the importance iterative and collaborative processes involving various levels of the organizational hierarchy and bi-directional information flow as key factors to effective strategic planning and resource allocation.

(Reaction) The value of using quantitative methods lies mostly on doing the process than on the output (i.e. figures) they produce (in the same way that the value of planning is on planning itself and not so much on the particular plan). The value is on the way these methods “encourage” the people using them to think systematically, analytically and critically. Within this framework, crucial information, experience and insights can be gained that otherwise would have gone unnoticed or undiscovered. It is critically important that decision-makers understand the meaning behind the figures. For in the end, selecting R&D projects should be based on sound (i.e. informed) judgment rather than on mechanical comparison of meaningless numbers.

B. “A Methodology for Research Project Selection”

(Summary) When considering R&D projects, funding agencies, both internal and external, require information that presents the projects' relation to the firm's financial and strategic goals. Often, multiple objectives and measures need to be considered. Therefore, these measures need to be combined in a consistent manner to compare projects. Furthermore, projects are often initiated in a bottom-up manner. For such projects, financial and strategic benefits are often not well identified and articulated (though they may be of great technical merit). Improvement Concept Selection Methodology (ICSM) is a project selection methodology developed to address these difficulties. ICTM allows management to effectively validate two things: 1) that a project is technologically and organizationally sound, and 2) that its benefits can be “proven”. ICTM uses the Enterprise Performance Management Methodology (EPMM) as its technology justification methodology. In EPMM, a comprehensive set of performance measures is used and integrated with activity-based management to asses a project’s impact on the enterprise-level. ICTM has for phases: Define Enterprise, Define Concepts, Plan and Conduct Validation Experiments, and Analyze Results. In the Define Enterprise phase, the enterprise's strategies and objectives are defined and linked. Based on them, a set of strategic metrics, considering both quantitative (e.g. NPR) and qualitative benefits (e.g. improved market share), is developed. Furthermore, a model of the enterprise's activities that might be impacted is also created. The developed set of strategic metrics and activity model are used to evaluate all projects to allow for a consistent comparison. In the Define Concepts phase, the impact of the project to the enterprise is envisioned and detailed. The interaction between specific components of the project and the specific activities are identified. It is also in this phase where risks are identified and plans to mitigate them are developed. In the Plan and Conduct Validation Experiments, experiments are designed and executed to evaluate the feasibility of a project and come up with data to assess its benefits. Finally, in the Analyze Results phase, the data obtained from the previous phase are analyzed and savings estimates are developed. Experiments are evaluated individually and synergistically (if applicable). The costs, savings and benefits are analyzed using an activity analysis matrix (from EPMM) and a strategic analysis matrix. An activity matrix maps project components (and their cost drivers and categories) to the enterprise activities they will impact. A strategic analysis maps project components to their quantitative and qualitative benefits.

(Reaction) A critical component in project selection is a rigorous, systematic methodology. Equally critical is the immersion of the individual participants in a collaborative process. Collaboration allows examination of projects from different angles and perspectives while a rigorous system keeps the perspectives directed and focused.