This paper develops a theoretical model to explain corporate divestment in the context of accounting-based valuation and provides empirical evidence to support the model's predictions. Building on Zhang's (2000) real-options-based equity value model, we develop a model to explain why firms with multiple business segments may have incentives in financial reporting to shift earnings from one segment to another to influence market valuation. Cross-segment earnings shifting, however, causes information asymmetry about segmental performance, which leads to market misvaluation. Divestment arises as a voluntary commitment by (some) firms to not engage in segmental earnings manipulation, with the aim of restoring valuation accuracy. Our theoretical analysis yields a number of testable implications. Consistent with our model's predictions, we find empirically that (1) divestment is preceded by an increased divergence in profitability between the divested and continuing segments of the divesting firm, (2) there are positive abnormal stock returns surrounding divestment announcements that are not dependent on increased expectations about future operating performance, (3) the magnitude of market revaluation increases with the profitability divergence between the divested and continuing segments, and (4) market revaluation is greater for more complex firms (in terms of having a larger number of segments and greater uncertainty facing investors).

Applying a real-options-based valuation approach, we develop and test a model that addresses the incremental value relevance of segment data beyond firmlevel accounting data. Prior studies (e.g., Zhang 2000; Biddle et al. 2001) show that equity valuation requires accounting data (in part) because accounting provides signals that guide capital investments underlying value creation. In this study, we establish that the usefulness of segment data beyond aggregate data relates to heterogeneity of investment opportunities across segments, caused by divergences of segment profitability and growth potential. Empirical results are consistent with the model's predictions. We also assess the magnitude of the valuation impact of segment information relative to that of firm-level information.

This study provides evidence of changes in how analysts generate stock recommendations after the SEC's approval of NASD Rule 2711 in May 2002, which introduced regulatory reforms to enhance the independence of analysts' research. We investigate the relations of analysts' stock recommendations with intrinsic value estimates (based on analysts' earnings forecasts relative to the stock prices, V/P) and with investment-banking-related conflicts of interest during the 1994–2005 period. We find a stronger relation between analysts' stock recommendations and V/P and a weaker relation between analysts' stock recommendations and conflicts of interest in the post-Rule period than prior to the implementation of the Rule. Moreover, the increases in the relation between stock recommendations and V/P after the implementation of the Rule are greater for the stocks recommended by analysts with greater potential conflicts of interest. Our findings suggest that the implementation of Rule 2711 has enhanced analysts' independence.

Prior studies show that equity value has convex relations with earnings and book value of equity, respectively, due to growth and adaptation options (Burgstahler and Dichev 1997a; Zhang 2000). However, these studies do not consider the role of institutions in affecting firms' ability to exercise growth and adaptation options. In this study, we investigate whether these convex relations vary with the degree of a country's economic freedom, which may influence the frictions and costs of exercising these options. We develop four hypotheses: In countries with greater economic freedom: (1) a firm's capital investment in response to profitability is greater; (2) the relation between equity value and earnings, given equity book value, is more convex; (3) the relation between equity value and equity book value, given earnings, is more convex; and (4) the relation between stock return and profitability change is more convex. Using the Economic Freedom of the World index from the Fraser Institute, we test our hypotheses with data from 30 countries during the 2000–2010 period. The empirical results are consistent with these hypotheses. The effect of economic freedom that we document is distinct from the effects of GDP level and growth, legal origin, law enforcement, investor protection, and quality of accounting standards. Our results suggest that greater economic freedom enhances equity value through more efficient management of investment options. Data Availability: Data used in this study are available from public sources.