Since the current financial reporting uses various measurement attributes for measuring assets and liabilities, it is often referred to as the “mixed measurement model.” This paper addresses the issues related to why and how different measurement attributes, notably fair value and historical cost, are used in financial statements. The cost-benefit analysis is adopted as a research methodology throughout this paper by formalizing as mathematical functions the preparer’s measurement cost and the users’ benefits of accounting information that the preparer produces. The cost-benefit analysis suggests that the mixed measurement model can be justified through the preparer’s economic incentives for minimizing cost. It also suggests that the cost-minimization solution in the marketplace would represent a consensus among the interests of the preparer and the users within a constraint in which only one set of accounting information is provided. This paper also identifies the situations in which providing two sets of accounting information is preferable.
We have explored the extent of accrual-based and real activities-based earnings management using data from family and non-family firms in Japan. Family firms are expected to have lower agency costs because family shareholders and management are more congruent in their pursuit of mutual firm goals and seek lower levels of earnings management. However, this collusion may lead to entrenchment and higher levels of earnings management, which becomes opaque to outside shareholders. A founding family is concerned with the reputation of their firm for sustained socioemotional wealth and family firms may conduct cosmetic earnings management to conceal bad news. We empirically assess the levels of earnings management and investigate whether the level will be lower or higher for family or non-family firms, and identify which method is more costly. The level of accruals and cost may vary among the family firm types; that is, whether or not shareholdings are large or the CEO is from the founding family. We find that the level of both accrual-based and real activity measures is lower for family firms. With cross-section regressions, we find that family shareholding increase the level of abnormal accruals management, whereas the family CEO decreases the level of abnormal accruals, but in both cases the amounts were not significant. We also find that family-related variables decrease the levels of real-activities earnings management. When we introduce economic measures related to the costs of earnings management, we find that Japanese family firms utilize accrual-based earnings management more often than real activities-based earnings management.