Until 1974, firms could choose, within GAAP, to capitalize or expense interest costs associated with capital expenditures. The predominant practice had been to treat interest as a period expense. However, in 1974, the Securities and Exchange Commission imposed a moratorium on further adoption of interest capitalization by non-regulated firms. This study empirically examines economic factors potentially influencing firms' decisions to expense or capitalize interest prior to the SEC moratorium. We hypothesize that the choice may be affected by (1) the existence of management compensation agreements tied to reported earnings, (2) debt covenant constraints, and (3) the political costs (for some firms) of reporting higher earnings. When compared to the control group, our findings are that (1) the frequency of explicit management compensation packages was not greater for the interest capitalization group, (2) firms with financial ratios closer to likely debt agreement constraints (on dividends, interest coverage, and leverage) tended to elect interest capitalization, and (3) other than the largest firms in the ‘politically sensitive’ petroleum refining industry, the larger firms were more likely to capitalize interest.

This paper is designed to provide additional evidence on the positive theory of accounting policy choice by combining individual accounting principles into firm income strategies. These strategies were the dependent variable in a probit analysis where the independent variables were size, management compensation, industry concentration ratio, systematic risk, capital intensity and the total debt to total asset ratio. The results indicate that four of these factors (size, management compensation, concentration ratio, and the total debt to total asset ratio) have a significant association with the choice of a firm's income strategy. This test provides strong evidence consistent with the positive theory of accounting standard setting/choice. We also present evidence that smaller firms and/or firms in less concentrated industries do not appear to make accounting policy choice decisions that are consistent with this theory