I construct a daily business cycle index based on quarterly GDP growth and textual information contained in a daily business newspaper. The newspaper data are decomposed into time series representing news topics, while the business cycle index is estimated using the topics and a time-varying dynamic factor model where dynamic sparsity is enforced upon the factor loadings using a latent threshold mechanism. The resulting index classifies the phases of the business cycle with almost perfect accuracy and provides broad-based high-frequency information about the type of news that drive or reflect economic fluctuations. In out-of-sample nowcasting experiments, the model is competitive with forecast combination systems and expert judgment, and produces forecasts with predictive power for future revisions in GDP. Thus, news reduces noise. Supplementary materials for this article are available online.