경영관행의 조직 내 확산과 편향적 조직학습 : 다국적 금융기업의 팀형성과 그 성과를 중심으로

Once an organization adopts a management practice, its members begin to learn how to implement the practice in their business units. Prior research has proposed two scenarios: (1) Tight coupling,where the organization successfully implements the formally adopted practice and integrates it with existing business routines, or (2) Decoupling, where the organization adopts but does not effectively implements the practice. A third scenario is also possible, where the management practice is initially implemented and diffused widely among organizational members / subunits, but winds down rapidly soon after a short period of implementation. While anecdotally this third scenario,which I term ‘faddish coupling’, seems to be a likely outcome for many organizations, little attention has been paid to how such a trajectory comes about. Building on organizational learning theory, this paper argues that biased organizational learning by organizational members and subunits results in faddish coupling of an organizational practice. Organizational subunits can learn from their own experiences and other units’ experiences, but such experiential and vicarious learning may be extremely biased, especially when members /subunits are under pressure to commit to company strategies and newly established organizational goals. When boundedly rational decision makers are faced with such strong pressures but not equipped with related knowledge about novel management practice, they tend to narrow their search scope down to a few salient and easily accessible activities such as the most recent high-performance experiences and examples, thereby carrying out organizational learning on the basis of positively-biased information. Organizational learning from a biased sample often produces positive sentiments and perceptions about the management practice, leading to its organization-wide diffusion. However, such biased learning, together with an elevated aspiration level, is likely to aggravate the quality of decision-making, drive below-aspiration practice performance levels, incur rapidly increasing frustration, and result in the sudden collapse of the adopted practice. Using data from cross-functional process improvement teams of a leading global financial service organization (under the pseudonym Global Financial) which adopted a total quality management program in the first quarter of 1997 and formally abandoned it in the last quarter of 2000, this paper tests several hypotheses regarding the processes and consequences of biased organizational learning involved in quality team formation in 174 organizational units operating in 77 countries. I evaluated the data against the following hypotheses regarding learning bias under conditions of strong organizational pressure:Hypothesis 1: A subunit’s team formation rate will be positively affected by the most recent team performance within that subunit. Hypothesis 2: A subunit’s team formation rate will be positively affected by the most recent team performance of reference units (geographically and operationally similar units). Hypothesis 3: A subunit’s team formation rate will be positively affected by the most salient team performance of reference units. But such biased learning may not last long. As the experience with team activities accumulates,organizational units become aware that biased learning brings about relatively poor outcomes,for example, low team performance and reduced conclusion of team activities. In hypothetical terms:Hypothesis 4: As the experience with team formation accumulates, biased learning from a selected sample will decrease. Hypothesis 5: Learning from a positively biased sample will decrease team performance. Hypothesis 6: Learning from a positively biased sample will decrease the probability of successful completion of team activities. Empirical analyses of quality team formation in Global Financial’s organizational subunits support all of the hypotheses. Organizational units’ learning appears to be biased toward the most salient team activities of their own and reference units, but such biased learning gradually tapers off as the experience with team activities within units accumulates. In addition, the findings show that biased learning has a negative impact on teams’ financial performance and their substantiality. The unexpected negative outcomes created a negative learning loop, which was followed by the rapid demise of the quality program in this organization. Finally, I discuss some implications for organizational learning theory and management studies and provide several possible future directions for organizational research. Traditional approaches to organizational learning have maintained an unjustifiable assumption that organizational learning almost always expands sound knowledge stock, enhances organizational capabilities, and improves performance. However, this paper suggests that researchers should reconsider this assumption and pay more attention to the negative effects of organizational learning. To reach a full understanding of how organizations learn and the consequences of organizational learning, it is important to establish a balanced view of organizational learning. The paper also suggests that future research should delve deeply into the area of so-called meta-learning (i.e. learning to learn), because when prior organizational learning does not deliver desired outcomes, organizations can change learning methods, information sources, and search scope.