An investigation of the accuracy of pre-implementation estimates required by SAB 74

Abstract

Staff Accounting Bulletin No. 74 (SAB 74) of the Securities and Exchange Commission (SEC) requires registrants to describe the impact of newly issued financial accounting standards not yet implemented by the company. In the case of accounting for postretirement benefits under Financial Accounting Standards Board Statement No. 106 ( FASB, 1990), many corporations complied with the provisions of SAB 74 by disclosing an estimate of the obligation anticipated from implementing the accrual method. The purpose of our study is to determine the format of pre-implementation estimates, the accuracy of these estimates, and the factors related to estimate inaccuracy. In addition, our study compares management forecasts with an estimation model based on cash payments and employee turnover. We found that management estimates took the form of either specific amounts or ranges. Much of the sample (71.4%) provided estimates that were in error by 10% or more. In addition, 70.5% of the companies underestimated the obligation for postretirement benefits. The variation between estimated and actual benefit obligations was related primarily to two factors: the size of the corporate benefit plan and the volatility of benefit costs. However, management estimates were, on average, more accurate than estimates generated by an external estimation model developed by Warshawsky et al. (1993, pp. 193–195).