Purchase, Pooling, and Equity Analysts' Valuation Judgments

We investigate whether analysts' common-stock valuation judgments are predictably affected by (1) different methods of accounting for business combinations and (2) the number of years that elapse after the business combinations occur. Numerous articles in the business press suggest that companies go to great lengths to avoid amortization of the purchase-method acquisition premium, either by structuring acquisitions to qualify for pooling treatment or aggressively allocating the acquisition premium to "in-process" research and development (IPRD) and immediately expensing it. Firms' reluctance to capitalize and amortize the acquisition premium often is based on their managers' belief that amortizing the acquisition premium impairs firm value, especially in years subsequent to the transaction. However, research to date neither refutes nor defends this belief. We report the results of an experiment in which 113 buy-side equity analysts participated. Consistent with our expectations, analysts' post-combination valuation judgments are lower when a parent company records and amortizes an acquisition premium in a purchase-method business combination and equivalently higher when the company either immediately expenses the entire purchase-method acquisition premium as IPRD or records the business combination using the pooling-of-interests method. In addition, in the case where the parent company records and amortizes an acquisition premium in a purchase-method business combination, analysts' stock-price judgments are significantly lower if the business combination occurred three years ago as compared to one year ago. We also test the FASB's proposal to report separate sub-totals (and related EPS figures) for pre-goodwill-amortization operating income, after-tax goodwill amortization, and net income. Our results suggest that the Board's proposed goodwill-reporting format mitigates the valuation difference between combinations that occurred one-year ago versus three-years ago. We discuss the implications of this study for users of financial accounting information and for accounting standard setters.