1990년대 이래 인적자원관리 시스템이 조직의 경영성과에 미치는 영향을 분석한 연구들이 활발하게 이루어졌으며, 그 동안의 연구결과에 따르면 고성과 인적자원관리 시스템이 조직의 경영성과에 통계적으로 뿐만 아니라, 경제적으로도 유의한 수준의 긍정적 영향을 미치는 것으로 나타났다. 그러나 고성과 인적자원관리 시스템 지표(index)를 계산하기 위해 선행 연구들이 채택한 방식, 즉, 고성과 인적자원관리 시스템의 구성요소라 할 수 있는 개별 제도들과 조직의 경영성과 사이에 선형적 관계(linear relationship)가 존재한다는 전제 하에 개별 제도들의 시행강도를 단순 합산한 가산적 방식(additive method)이 적정한지에 대한 검토가 필요하다. 왜냐하면 일부 고성과 인적자원관리 시스템의 구성요소로 알려진 일부 제도들은 조직의 경영성과와 비선형적 관계(non-linear relationship)를 가질 수 있음이 밝혀졌기 때문이다. 본 연구는 고성과 인적자원관리 시스템 지표 계산방식으로 활용 가능한 프로파일 방식(profile deviation method)을 적용하여 고성과 인적자원관리 시스템과 조직의 경영성과 사이에 인과관계가 존재함을 확인하였다. 뿐만 아니라, 일부 고성과 인적자원 관리 제도가 조직의 경영성과와 비선형적 관계를 가지고 있을 경우 가산적 방식보다 프로파일 방식을 사용하여 고성과 인적자원관리 시스템 지표를 계산할 때 조직의 경영성과에 미치는 한계효과를 더 잘 포착할 수 있음을 확인하였다. 결론과 토의 부분에서 본 연구의 의의와 한계를 제시하였다.
The effects of human resource management (HRM) system on firm-level performance have been extensively examined since the early 1990s, and the effect sizes were reportedly very promising. Huselid and Becker (2000), based on their research using four national surveys over 2,000 firms, concluded that "the effect of a one standard deviation change in the HPWPs is 10-20% of a firm`s market value." In addition, according to Combs et al.`s (2006) meta-analysis, the corrected effect size of the HPWPs on firm performance was .20, which means that "a one standard deviation increase in the use of HPWPs translates, on average, to a 4.6 percentage-point increase in gross ROA." However, in many previous studies focusing on the level of system or bundle of HR practices, it has been implicitly assumed that the relationship between each high performance work practice (HPWP) and firm performance was positively linear, which means that the more intensively does each HPWP within the bundle get implemented, the better firm performance is expected. The assumption has been reflected in the way of researchers` calculating a high performance HR system index, in which the extents to which individual HPWPs are implemented in the workplace are additively summed up. However, this method is challenged by research findings that some motivation-enhancing HPWPs could have non-linear relationships with firm performance. Psychological research suggests that excessive rewards can result in a decline in individual performance. One mechanism that can produce a perverse relationship between motivation and performance is related to increased self-consciousness (Baumeister, 1984; Langer & Imber, 1979; requoted from Ariely et al., 2005). Increasing self-consciousness due to a strong cash incentive can cause people to consciously think about the task, shifting control from ``automatic`` to ``controlled`` processes that are less effective. Another mechanism by which increased motivation is also likely to have a negative effect on performance relates to a general focus of attention (Easterbrook, 1959; requoted from Ariely et al., 2005). Attentional focus can be detrimental for tasks that involve insight or creativity, since increased motivation tends to narrow individuals` focus of attention to the way they are accustomed to, although creativity and insight require drawing unusual connections between elements. McGraw & McCullers (1978) provided support for this mechanism by showing that the introduction of monetary rewards for tasks that involved problem-solving had detrimental effects on performance. In addition, large incentives intented to increase extrinsic motivation can do harm to intrinsic motivation (Deci, Koestner, & Ryan, 1999). Research examining pay dispersion also suggests that pay differences among employees in the same rank can generate both positive and negative effects within the workplace. With higher levels of pay dispersion, greater potential exists for employers to motivate employees through the use of financial rewards (Cadsby, Song & Tapon, 2007). However, pay dispersion also creates a potential for competition among employees to interfere with cooperation (Yang & Klaas, in press). Further, to the extent that the legitimacy of managerial judgments regarding individual performance remains questionable, greater pay dispersion has the potential to lead to higher levels of perceived inequity. Similarly, Shaw, Gupta & Delery (2002) found that pay dispersion among employees negatively affected organizational outcomes when there were high levels of interdependence and also when the firm lacked systematic means by which to measure employee contributions. Yang & Klaas (in press) also found that a horizontal pay dispersion practice has an inverted U-shaped relationship with firm performance among Korean firms. The above research findings tell that the additive way of creating a HPWP system index based on the assumption that the relationship between each HPWP and firm performance is positively linear needs to be reexamined. This study designed to figure out the causal relationship between a high performance HR system and firm performance by adopting an alternative way of indexing a high performance HR system, that is, a profile deviation method. A profile deviation method in this context sets up an ideal profile of high performance HR system and then calculate a firm`s HR system index based on the extent to which its HR system deviates from the ideal high performance HR system. This profile deviation method can be flexibly used even under the condition that the relationship between a certain HPWP and firm performance is not linear. The hypothesis was tested with two data sets obtained from a government sponsored survey, which has been conducted every two years since 2005. The sample was drawn from the firms with more than 100 employees in the corporate database maintained by the Korea Information Service and firm performance data for relevant years which were obtained from the KIS corporate database were combined with the survey data. In 2005 data, 409 firms were included, while in 2007 data, 419 firms were included in the sample. In this study high performance HR system indices were drawn from two different methods, that is, an additive method and a profile deviation method for a comparison purpose. As far as the research design was concerned, a ``predictive`` design, where the relationship between HR practices and post-performance is explored, as compared to ``post-predictive`` design, where the relationship between current HR system and past performance is explored, or ``contemporaneous`` design, where the relationship between the contemporaneous HR system and performance is explored, was adopted with past firm performance controlled. The results show that a causal relationship between high performance HR system and firm performance exists. Meanwhile, the profile deviation method was found to be more powerful in detecting existing effects than the additive method when some high performance HPWPs have a non-linear relationship with firm performance. The result implies that the profile deviation method is more robust than the additive method in detecting the effects of a high performance HR system on firm performance. The implications and limitations of the study are discussed in conclusion section.
