We use the daily internet search volume from millions of households to reveal market-level sentiment. By aggregating the volume of queries related to household concerns (e.g. “recession”, “unemployment” and “bankruptcy”), we construct a Financial and Economic Attitudes Revealed by Search (FEARS) index as a new measure of investor sentiment. Between 2004 and 2011, we find increases in FEARS lead to return reversals: although high FEARS are associated with low returns today, they predict high returns over the following two days. In the cross-section of stocks, the reversal effect is strongest among stocks which are attractive to noise traders and hard to arbitrage. FEARS also coincide with excess volatility and predict daily mutual fund flow. When FEARS increase, investors are more likely to pull money out of equity mutual funds and put it into bond funds. Taken together, the results are broadly consistent with theories of investor sentiment.