We examine the credibility of synergy guidance in 12,176 U.S. mergers and acquisitions (M&A) announced between 2004 and 2021. Using press releases and conference call transcripts, we identify qualitative and quantitative synergy guidance and analyze its determinants, market reactions, and post-acquisition realizations. Synergy guidance is more common in larger, stock-financed deals, public-target acquisitions, related-industry transactions, and among acquirers with greater analyst coverage and prior synergy disclosure experience. Litigation risk deters disclosure, especially numeric guidance. Investors respond favorably to both textual and numeric guidance at announcement, with larger reactions to more intensive discussion and higher numeric estimates. However, post-acquisition outcomes suggest that disclosed expectations are often not realized. Synergy disclosers experience more frequent and larger goodwill impairments and worse post-merger operating performance, while larger numeric estimates predict larger impairments, lower long-run stock returns, and downward revisions to synergy expectations. This impairment association is attenuated when acquirers have greater forecasting experience and targets operate in richer information environments, consistent with learning and information quality constraining overoptimism. Exploiting the 2013 U.K. Takeover Panel Rule 28 changes, we find that numeric guidance in U.K.-target deals predicts fewer subsequent goodwill impairments than comparable U.S.-target deals. Our findings inform ongoing policy debates by highlighting a disconnect between investors' favorable announcement reaction to synergy guidance and its limited long-run realization.

Separating operating from financial effects helps capture firm fundamentals, as these two types of risk tend to offset each other in the cross-section. We revisit the Fama-French six-factor model (FF6; 2018) by recomputing factors using operating characteristics while separately accounting for financial leverage. Our six-factor model subsumes the original FF6 model and outperforms the q-factor model (HXZ; 2015) in explaining a large set of documented anomalies. The improvement in capturing operating risk primarily stems from using return on net operating assets and growth in net operating assets as measures of profitability and investment, respectively. The enterprise book-to-market ratio becomes redundant once we consider the other operating characteristics. Additionally, after we control for operating effects, leverage is positively associated with returns, as predicted by theory, suggesting a renewed role for a leverage factor in asset pricing.