원가의 비대칭성에 관한 연구

Abstract

Factors that influence cost behavior and the role of managers in controlling costs are central questions in management accounting. Understanding the sources of cross-sectional variation is an important way to test alternate theories and gain insight into how managers’ decisions influence realized costs. In light of recent debate concerning whether costs are sticky (Anderson, Banker and Janakiraman 2003) or are a data artifact (Anderson and Lanen 2007), we explore three dimensions expected to influence the stickiness of costs. First, using ownership as a proxy for organization level incentives and constraints, we examine whether the stickiness of costs varies predictably across organizations. Next, we examine two factors that likely influence withinorganization variation in the stickiness of costs. We posit that managers are less likely to trim costs in functions related to the organization’s core competency and that the nature of the resource affects its adjustment costs. Accordingly, we investigate whether cost stickiness varies across groupings of resources by function (e.g., patient care versus services within a hospital) or cost category (e.g., capital versus labor versus other). In light of recent debate concerning whether costs are sticky (Anderson, Banker and Janakiraman 2003) or whether stickiness is a data artifact (Anderson and Lanen 2007), we explore dimensions expected to influence the stickiness of costs. We examine whether stickiness varies predictably across organizations according to organization type, and within-organizations based on relation to the organization’s core competency and nature of the resource. We find strong evidence of stickiness, overall. Moreover, we find that costs of core activities react less to reduced activity relative to costs for non-core activities. However, we do not find expected variation across more specific expense categories such as labor and other services. Similarly, while we find differences in stickiness across organizations, the evidence does not produce a strong or coherent picture. Collectively, our findings suggest that we either need large samples or aggregate data to reliably detect stickiness in costs, even in settings where we expect stark differences. We conjecture that contextual knowledge gained from field studies might be indispensible in constructing tests that can provide further insight into managerial choices relating to cost management.