We examine whether fair value measurements enhance analysts’ forecasting accuracy.1 Fair value measurements may augment forecasting, providing more timely data than historical cost measurements. However, such measurements may lack reliability and the presence of fair value items may increase the volatility of earnings, making the forecasting task more difficult. Prior studies have provided some evidence on the relationships between fair value measurements, firm value, and cost of capital, suggesting that these measurements are relevant to investors and creditors (Arora, Richardson, & Tuna, 2014; Song, Thomas, & Yi, 2010; Riedl & Serafeim, 2011; Goh, Li, Ng, & Ow Yong, 2015; Barth, 1994; Barth, Beaver, & Landsman, 1996; Barth, Hodder, & Stubben, 2008; Barth & Taylor, 2010; Barth, Ormazabal, & Taylor, 2012; Blankespoor, Linsmeier, Petroni, & Shakespeare, 2013; Graham, Lefanowicz, & Petroni, 2003; Carroll, Linsmeier, & Petroni, 2003; Venkatachalam, 1996). Yet, limited evidence exists on the relationship between fair value measurements and the accuracy of analyst earnings forecasts (see Magnan et al., 2015). We expand research in this area by examining the relationship between fair value measurements and analysts’ information environment using a sample that includes financial and non-financial firms and comparing the relevance of these measurements in times of economic stability versus times of economic distress. Developing a comprehensive understanding of the usefulness of fair value measurements is important as it can inform accounting standard setters and regulators on this issue. Over the past 25 years, the Financial Accounting Standards Board (FASB) has expanded the use of fair value measurements to include items such as derivatives and hedges, employee stock options, financial assets, and goodwill impairment testing. A significant standard in this area was Statement of Financial Accounting Standards (SFAS) No. 157, Fair Value Measurements. SFAS No. 157 established a framework of fair value measurement and required fair value measurements to be disclosed by levels (Level 1, 2, and 3), with Level 1 having the highest measurement certainty and Level 3 having the lowest level of measurement certainty Because Level 3 measurements inputs that are often not observable by investors, they are subject to greater estimation errors and biases, potentially causing them to be less reliable and create more severe information asymmetry between mangers and investors.2 Our study joins research streams that investigate both the usefulness of fair value measurements and the information used in the formation of analyst earnings forecasts. Analytical studies of analysts’ behaviors provide models that illustrate the relation between information quality and the characteristics of earnings forecasts (Diamond, 1985; Kim & Verrecchia, 1997; Barron, Kim, Lim, & Stevens, 1998). These studies suggest that more useful disclosures result in more accurate and less disperse earnings forecasts. Building upon these analytical models, empirical studies have employed the characteristics of analyst forecasts as a proxy for the quality of measurements (see, for example, Byard, Li, & Yu, 2011). We expand these studies by exploring the impact of fair value measurements across firms of different types under different economic conditions.