options are not capitalized. Thus, the presentation effects we observe appear to be linked in part to the weight that participants give to nonfinancial factors arising from the contractual lease arrangement. To further explore this relationship, we conduct a mediation analysis.22 In this analysis, we use
disaggregation of the lease liability as the independent variable, participants’ lending decision as the dependent variable, and the importance assigned to nonfinancial factors as the mediating variable (Baron and Kenny 1986).23 As shown in Figure 3, the importance assigned to nonfinancial factors fully mediates the relationship between disaggregation and lending decisions. Thus, when lenders place importance on nonfinancial factors, such as exposure to rental rate fluctuations and operational flexibility, they are more likely to lend to the lessee firm, regardless of how the lease obligation is presented. However, disaggregating the capitalized renewal periods increases the extent to which lenders emphasize these nonfinancial factors, which helps explain the increase in lending in this condition.
Evidence on Lending Decisions when Disaggregation Occurs Only in the Notes
Our capitalized and disaggregated condition (Condition 3) breaks out the fixed and the optional components of the lease liability in both the financial statement and the notes accompanying the financial statement. We chose this approach to increase the likelihood that participants would attend to the disaggregation manipulation, allowing us to create a more powerful design for testing whether disaggregation of the capitalized lease obligation increases lending rates. As a result, however, we are unable to determine whether participants are reacting to the disaggregation of information in the notes, in the financial statements, or both.
While a number of prior studies examine the issue of recognition versus disclosure, only a few directly address the issue of information location, holding constant the decision to recognize an item in the financial statements. For example, Clor-Proell et al. (2010) show that disaggregating fair value information on the face of the income statement can affect investors’ information processing, but their design does not manipulate the location of this disaggregated information. In contrast, Bloomfield et