|
Article Excerpt
1. Introduction
Understanding the value of a product development project is essential to the scientific management of the product development process. The value of a project to a firm depends not only on the project's properties but also on those of the other projects the firm is developing. This is due to interactions with the other projects in the firm's same development portfolio. Understanding these portfolio-level project interactions is central to a firm's choice of project portfolio and development capacity (Loch et al. 2001a, Kavadias and Loch 2003). Decision support models for portfolio choice have provided analytical models of these interactions (cf. Ding and Eliashberg 2002, Loch and Kavadias 2002). In this study, we empirically investigate the structure and significance of these interaction effects.
We use the natural experiment of a product development failure to estimate the value of an individual project. We design an event study around the failure of a late-stage development project. This event study gives us a metric of the change in the firm's value (as measured by the stock markets) due to the failure of a development project with all other factors affecting the firm's value being held constant (MacKinlay 1997). This change in firm value is an empirical measure of the value of the failed project to the firm. We then explain the variance in the value of all failed projects in our sample based on the interactions of the project with other projects in the portfolio. Specifically, we investigate how the value of a project to a firm may depend on the presence of other projects in the firm's portfolio that address the same customer need or utilize the same development resources.
The specific context of our empirical examination is the pharmaceutical industry. New product development in the pharmaceutical industry is regulated and thus, proceeds along a series of well-defined steps illustrated in Figure 1. Drug development starts with an investigation of the chemical and biological properties of a compound in the lab (basic research), followed by animal trials (preclinical studies) and, finally, three stages of clinical trials or trials in human subjects (phase I, II, and III). Our study is designed around the failure of development projects currently undergoing phase III clinical trials, which is the final stage in the development process, where the efficacy and safety of the drug is investigated in a large sample of patients. Common causes for failures at this stage include adverse side effects of the drug and harmful interactions with other drugs. For a detailed description of the drug development process, the reader is referred to Pisano and Rossi (1994) and Girotra et al. (2004).
[FIGURE 1 OMITTED]
The pharmaceutical industry presents an ideal domain of enquiry for our study. It is a large industry where product development is central. Although the product development process closely resembles the classic stage-gate development process prevalent in most industries, the role of regulators in the later phases of the drug development process significantly simplifies the empirical design of our study. The different stages in the product development process are clearly and uniformly defined by the regulator, and all pharmaceutical firms must pass their development projects through the same development stages. This allows us to identify projects at the same stage of development across different firms in the industry. The results of each stage of the development process are public knowledge. This allows us to create our data set of product development failures from public sources. Finally, the late-stage product development portfolio of each firm is public knowledge. Thus, when the stock markets value failures, they have the information on the portfolio-level project interactions which we are investigating.
We find empirical evidence for two portfolio-level project interactions. First, we find that the impact of the failure, which is our measure for the value of the project to the firm, is smaller when the firm is developing other projects for the same market as the failed project. When the firm is developing multiple projects for the same market, the failure of one of them does not preclude the firm from earning sales in that market. Thus, the marginal value of any of the multiple projects is smaller than the value of a single project being developed for the market.
Second, we find evidence that the value of a compound or the impact of a failure is smaller if the firm has more projects than the anticipated number in its portfolio, which require the same resources as those used by the failed project (even if the other projects are not "backups" for the same market). A failure leads to the freeing up of resources shared by the failed and other projects. These freed-up resources can be redirected to other projects, which may then be brought to the market sooner than they would be if there was no failure. Thus, failures in portfolios with more than the anticipated number of projects that utilize the same resources as the failed project lead to the acceleration of the other compounds in the portfolio and have a smaller impact.
This study enhances our understanding of portfolio-level project interactions. We build and validate a theory on the financial effect of these interactions. We find that these interactions significantly alter the value of a development project to a firm and are thus crucial to portfolio choices. Our empirical evidence also allows for a critical examination of the existing analytical literature on portfolio and capacity choices with respect to the modeling of project interactions. This can help us understand the reasons behind the limited application of this literature in practice (Loch et al. 2001a, Loch and Kavadias 2002, Shane and Ulrich 2004) and inspire new improved analytical models, which take into account the empirical regularities that we find. Finally, our results also provide a data-driven model that aids in valuing individual projects in the context of the product development portfolio of a firm. This is useful in valuing development projects available for in-licensing and comparing alternative development projects.
2. Prior Literature
Two streams of academic work are relevant to this study: (1) research on portfolio choices and (2) research on the financial impact of product development outcomes. An established body of literature in operations research attempts to provide optimal product portfolio decisions. Initially, optimization models were developed in a static and deterministic setting, with the decision modeled as one-shot choice under complete information, often with a mathematical programming formulation (see e.g., Lucas 1971). More recent work has emphasized the stochastic, dynamic, or process nature of the problem and has analyzed capacity and congestion effects (Loch and Terwiesch 1999) as well as strategies for search and information gathering (Loch et al. 2001b, Dahan and Mendelson 2001).
Portfolio-level project interactions are central in many of the contemporary models on portfolio selection. Loch and Kavadias (2002) present a dynamic model of portfolio selection within a budget constraint, taking into account multiple project interactions, including those arising out of shared markets in a general setting. In their study on the number of development approaches to pursue for a given market, Dahan and Mendelson (2001) model the interactions between projects of different quality that target the same market. Ding and Eliashberg (2002) investigate the number of development approaches to pursue for a given market in a staged development process. They build an analytical model of the interactions between projects targeting the same market. In their model, unlike Dahan and Mendelson (2001), all successful projects are assumed to have identical quality. In our paper, we empirically examine interactions similar to those investigated by Ding and Eliashberg (2002).
Adler et al. (1995) build a model of project interactions due to shared development resources. Using a development project as their unit of analysis they find that if development resources are stretched, the project completion times are longer. In contrast to Adler et al. (1995), we take an empirical approach and study the effect of shared development resources at the portfolio level. We examine the impact of one project on other projects in the portfolio. Further, we find the impact of these interactions on the financial value of the project as opposed to the development lead time.
Multiple studies have focused on the impact of product development events on financial value, notably Hendricks and Singhal (1997) on the impact of product development delays. They find significant negative stock returns associated with the announcement of product introduction delays. Industry competitiveness and the firm's degree of diversification influence the size of this impact. Chaney et al. (1991) and Chaney and Devinney (1992) study the stock market reaction to announcements of new products across a wide range of industries. Bayus et al. (2003) study the impact of new product introductions in the personal computer industry on profit rate, profit rate persistence, and asset growth. Robertson et al. (1995) and Chen et al. (2005a) study the impact of new product announcements on competing firms. Chen et al. (2005b) examine the effect of product introduction delays on industry rivals. Sharma and Lacey (2004) compare the impact of pharmaceutical successes and failures on firm value. This body of work quantifies the impact of these product development events. Further, they explain the variance in the financial impact of product development with product or industry properties, but not the portfolio.
We build on the rigorous methodologies developed in this literature to empirically value projects. However, in contrast to this literature, we relate the impact of the product development outcomes (our measure for...
|
