|
Article Excerpt
1. Introduction
The question of which innovation programs to pursue is critical to a firm's success and is often cited as a key competitive dimension (Roussel et al. 1991, Wheelwright and Clark 1992, Cooper et al. 1998). Indeed, the first step in transforming strategy from a hopeful statement about the future into an operational reality is to allocate resources to innovation and new product development (NPD) programs in a portfolio. Determining an appropriate level of funding for innovation programs is difficult, particularly when programs conflict in terms of rewards, risks, and resource requirements. Success then requires a difficult balancing act: short-term benefits accrued through incremental improvements to existing products versus long-term benefits achieved through radical, new-to-the-market, or new-to-the-world products and services (Tushman and O'Reilly 1996).
Further complicating matters is the fact that strategy formulation and resource allocation typically occur in a top-down manner within the firm (Loch and Tapper 2002, Anderson and Joglekar 2005, Kavadias and Chao 2008). Strategies outlined by senior executives define the operating environment for subordinate levels of decision making. Consider the following example, typical of any large corporation: Senior executives set broad goals regarding corporate strategy (e.g., five years from today, 50% of revenue will come from new products; over the next year, costs must be reduced by 5%). Managers are responsible for transforming that strategy into a reality. To do so, they make decisions with respect to broad programs within the portfolio (e.g., halt next-generation technological development and embark on manufacturing process improvements).
A rich history of scholarship considers the resource allocation problem and the incentive issues created by organizational hierarchies. Practice has evolved, however, to incorporate various rules not often considered by these research streams. (1) These rules often determine which level in the organization's hierarchy has the authority to make decisions (Aghion and Tirole 1997, Baker et al. 1999). One particular rule that has a profound effect on NPD processes is that which determines funding authority for innovation initiatives. We say that NPD funding is fixed when a manager does not have authority over the manner in which the budget is created. Alternatively, we say that NPD funding is variable when the manager has the authority to use revenue derived from existing product sales to fund NPD effort. In practice, managers often claim they need authority over NPD funding (a variable budget) to effectively manage resources. Despite this claim, firms seem to follow various policies with respect to funding authority, incentives, and resource allocation strategy (Christensen and Anthony 2005).
The hierarchical structure of decision making regarding resource allocation and NPD portfolio management, the prevalence of authority and incentive issues, and the fact that practice varies so widely with respect to these matters prompt us to ask the following question: How does funding authority impact incentives, resource allocation, and NPD portfolio strategy? In addition to funding authority, we also consider typical agency issues that result from the organization hierarchy. Senior management (the principal) determines incentives in the presence of noisy information regarding the resource allocation decisions of a manager (the agent). The agent is responsible for resource allocation decisions, and she is held accountable for those decisions. We borrow from a seminal model in the economics literature to account for the multiple innovation opportunities available to the agent (Holmstrom and Milgrom 1991), and we characterize accountability through the agent's career concerns (Zwiebel 1995, Holmstrom 1999, Siemsen 2008).
Our results indicate that funding authority has significant implications for both the structure of incentives and the allocation of resources. When managers have funding authority, their incentives for innovation are greater and they invest more in both existing product improvement (incremental effort) and new product development (innovative effort). However, variable funding may have an unintended side effect--it induces the manager to focus on existing product improvement to a greater degree than on new product development, and the relative balance in the NPD portfolio shifts toward incremental innovation. This effect is so robust that it takes place even when the manager's actions are perfectly observable. When actions are not perfectly observable, we uncover a form of substitution between wage incentives (explicit incentives) and the manager's career concerns (implicit incentives). Traditional wage incentives are reduced as the manager's career concerns become more significant. The lower wage incentives result in lower effort and, subsequently, lower joint surplus.
2. Related Literature
There is extensive literature on organization theory that focuses on the link between structure and performance (Galbraith 1977, Mintzberg 1980, Ouchi 1979, Eisenhardt 1985, Tushman and O'Reilly 1996). Much of this work is aimed at understanding the physical structure of work teams or business units (e.g., matrix, functional, or decentralized structures). Although structure is an important determinant of success, organization theorists note that authority and control are also important elements of organization design (Ouchi 1979, Eisenhardt 1985). Two decades of scholarly work take note of this observation (Lewin and Minton 1986, Barley and Kunda 1992, Kumar and Seth 1998, Bate et al. 2000). Important themes common to this body of research are the implicit and explicit incentives that drive action.
Incentive mechanisms have long been within the purview of researchers in economics. Grossman and Hart (1983) formalize the nature of compensation and explicit incentives, whereas Holmstrom (1999) formalizes the concept of career concerns as implicit incentives. Gibbons and Murphy (1992) consider the interplay between compensation (explicit incentives) and career concerns (implicit incentives) when the agent makes a single resource allocation decision. Kaarboe and Olsen (2006) extend the work of Gibbons and Murphy to a multitask setting. Aghion and Tirole (1997) discuss the tension between formal and real authority in organizations, and Baker et al. (1999) show that informal authority is beneficial in settings characterized by repeated interactions. We build upon this work in that we consider resource allocation between multiple innovation initiatives in the presence of implicit and explicit incentives, and we add the practical element of funding authority.
Researchers have studied the impact of incentives on NPD processes. Lambert (1986) models an agency problem in which the manager must be induced to select either a safe or a risky project. Our work considers the simultaneous balance between safe and risky efforts. Holmstrom (1989) considers a portfolio of efforts and shows that incentive costs associated with an individual task depend on the portfolio of tasks that a firm undertakes. Pollack and Zeckhauser (1996) study a situation in which a central agency must allocate a (fixed) budget between various business units. We extend this theory to situations in which the manager may have funding authority. Siemsen (2008) studies the design decisions made by engineers in the presence of career concerns. He finds that noisy signals with respect to capabilities may lead to over-design. We also consider a form of career concerns, and our link to fiscal responsibility is explicit.
A related stream of research exists in the area of corporate finance. This research considers the capital budgeting (resource allocation) problem in conjunction with issues related to incentives. Early work in this area focused on understanding the drivers of organizational slack (overfunding) and resource rationing (underfunding) for investment opportunities. Studies by Antle and Eppen (1985), Harris and Raviv (1996), and Bernardo et al. (2001) fall under this category. They model a single investment opportunity in a single-period setting. More recently, Dutta (2003) looks at the issue of managerial retention when the manager has private information about a (single) investment opportunity and he can pursue the project inside the firm or as an outside venture. Most closely related to our work is that of Stein (2002). He studies which type of organization form (decentralized versus hierarchy) is preferred depending on the portfolio of investments available to the firm. From a conceptual viewpoint, we differ from previous research in that we consider the dynamic allocation of resources between multiple projects in the NPD portfolio. Furthermore, we explicitly introduce the notion of fixed versus variable NPD funding and career concerns, which, to our knowledge, have not been addressed in prior work.
Our efforts enrich the literature along two important dimensions. First, we explicitly account for the realistic notion that NPD funding is not always fixed; rather, it may be determined by the authority given to a manager and his ability to generate revenue. Second, we analyze both explicit and implicit incentive mechanisms within the context of innovation. Because we include funding authority, incentives, and resource allocation decisions in one model, we are able to uncover the interaction between important processes that impact firm performance.
3. A Model of Incentives and Resource Allocation
Our model is best suited to a divisionalized firm in which a senior manager or CEO oversees multiple operating units, divisions, or groups, each of which has profit and loss responsibility. We focus our attention on the interaction between the CEO (the principal) and the senior decision maker (the agent) within one of the business units. Consider then a principal-agent relationship in which the agent makes a choice regarding resource allocation between incremental and radical NPD programs over a portfolio review cycle t [member of] [0, T]. The length of the portfolio review cycle may be as...
|
