What factors compel foreign subsidiary expatriate managers to continue investing in projects showing pessimistic economic prospects? Is that escalation tendency mainly attributed to the cognitive ability of individuals or can it also be attributed to incentive structures of individuals? Will the individual-level cultural values of decision makers affect the strength of investment escalation intention in a given situation? Efficient resources allocation decisions are essential for the success of a multinational corporation (MNC), which is under stiff global competition (Bartlett & Ghoshal, 1989). From the standpoint of an MNC's headquarters (HQ), investment projects in a certain host country are a part of the MNC's investments portfolio (Hitt, Hoskisson & Kim, 1997; O'Donnell, 2000). Thus, when the previous investments ultimately turn out to be sunk costs, an MNC should terminate further investments on the focal previous investment projects according to economic rationality (Dixit & Pindyck, 1994).
However, previous studies reveal that this economic rationality criterion does not always hold in actual situations (Arkes & Blumer, 1985; Bazerman, 1984; Garland, 1990; Schaubroeck & Davis, 1994; Staw & Ross, 1978). In many cases, decision makers tend to keep committing resources to a series of seemingly unprofitable investments. The existing literature calls the phenomenon "escalation of commitment to a failing course of action" (Brockner, 1992; Staw, 1981).
This escalation of commitment to unprofitable investment projects may particularly matter in the MNC context for the following reasons. First, foreign subsidiaries of an MNC are geographically and culturally more distant from the HQ than are domestic subsidiaries (Gong, 2003). These cultural and geographic distances incur additional costs that might be smaller in domestic operations. One of the typical types of costs incurred by institutional distances is agency cost (Eisenhardt, 1989; Jensen & Meckling, 1976). In order to reduce agency costs in foreign subsidiaries, the HQ tends to rely on managers expatriated from the home country (Gong, 2003). However, for various reasons, expatriate managers themselves may behave opportunistically as agents (Yan, Zhu & Hall, 2002), in which case, the parent firm (i.e., HQ in home country) of foreign subsidiaries will not be able to effectively monitor opportunistic behaviours of expatriate managers, such as additional investment decisions on unprofitable projects that have already incurred sunk costs.
The escalation of commitment may be particularly important for foreign subsidiaries in an MNC because expatriate managers, even when they are psychologically loyal to the parent firm, are exposed to cognitive biases that impede accurate project evaluations. One of the critical challenges that foreign subsidiary expatriates face is not about what to do in one-time decision making, but about "the fate of an entire course of action" (Staw, 1981). The ultimate outcome of a series of prior investments is often realized throughout the long-term investment horizon (Staw & Ross, 1978). This sequential nature of investments in a host country can make expatriates perceive their previous investment decisions not as a failing course of action. In addition, even if the expatriates perceive their previous investment decisions as unprofitable, they might perceive additional investment decisions as loss recovering commitments (Davis & Bobko, 1986). In such a situation, decision framing significantly affects the risk-taking propensity of the focal decision maker's investment decisions.
Except for a few studies (Harrison & Harrell, 1993; Sharp & Salter, 1997), the existing literature has understudied the effects of incentive structure and framing effect on investment escalation. This paper therefore aims to contribute to the literature by proposing four propositions that examine the effects of incentive structures and framings. In doing so, this paper provides theoretical and managerial implications about the motivational and cognitive aspects of escalation of commitment in the MNC subsidiary context. In addition, this paper can contribute to the literature by explicitly examining the moderating roles of individual-level cultural values, which are critical factors that have not received much attention in previous studies, due to the dominance of national cultural effects in the existing cross-cultural management literature.
THEORY AND PROPOSITIONS
The escalation of commitment to a failing course of action refers to "the tendency for decision makers to persist with" failing resource commitments in terms of economic efficiency (Brockner, 1992). The existing literature has explained the causes of escalation in terms of psychological, social and organizational determinants (Staw, 1997). Compared to the "lack of ability" aspect in evaluating project economics, one of the unfilled research gaps is the "motivation" aspect of escalation. Although self-justification theory provides a solid rationale for why individuals need to justify their previously committed decisions (Staw, 1997), this justification need has mainly been treated as a psychological determinant rather than as an opportunistic behaviour by individual organizational members as agents (Staw, 1997).
Incentive Structure in Expatriate Career Path
Strategic human resource management literature implicitly assumes that expatriates can protect the best interest of the parent firm from the agency costs incurred by locally hired staffs in foreign subsidiaries (Gong, 2003). However, this may not be always the case. Expatriate managers themselves may possibly behave opportunistically, when the pre-conditions of agency situations are met. Proposed research framework explained in Figure 1.
Agency theory explanations on escalation of commitment point out that both the misalignment of interests and the information asymmetry between principals and agents are preconditions of 'adverse selection' situations in foreign subsidiaries (Harrison & Harrell, 1993; Kirby & Davis, 1998). Within an MNC context, the parent firm can be a principal and foreign subsidiary managers can be agents of the parent firm.
First, information asymmetry needs to be present in order for agents to behave opportunistically. In other words, if the principal can also know that the project is failing, it is "in the best interests of the agent to discontinue a failing project" (Harrison & Harrell, 1993). Therefore, effective monitoring of foreign subsidiary managers by the parent firm could effectively safeguard escalation agency hazards. However, as previous studies (Gong, 2003; O'Donnell, 2000) point out, the cultural and geographic distance between the parent firm and the foreign subsidiary hinders the effective monitoring by the principal. That is, subsidiary managers could access private information to assist in evaluating the eventual fate of previously committed investments (Sharp & Salter, 1997). In this sense, the monitoring effect suggested by Kirby & Davis (1998) will be weak in the foreign subsidiary context.
Second, interest misalignment is necessary in order for agents to behave opportunistically. (Yan et al., 2002) illustrate how the long-term career horizon of expatriate managers can lead to goal conflicts between the principal and the agent. According to (Yan et al., 2002), regardless of the nature of the employment contract, when the principal and the agent have asymmetric perceptions of it, either agent opportunism or principal opportunism can arise.