Carhart's 1997 Study
Mark Carhart's landmark paper "On Persistence in Mutual Fund Performance," published in the Journal of Finance in 1997, is the most cited study on this question. Carhart examined a comprehensive dataset of US mutual funds from 1962 to 1993 and asked: after controlling for the momentum factor (which could mechanically create apparent persistence by rewarding funds that happened to hold last year's winning stocks), is there evidence of genuine alpha persistence?
The answer was essentially no. After removing the momentum effect, there was virtually no evidence of persistent alpha in mutual fund performance beyond one year. Top-quartile funds in one period were no more likely to be in the top quartile in the following period than chance would predict. The only meaningful persistence Carhart found was negative: poor performers were somewhat more likely to keep underperforming — largely because high expense ratios are a persistent and reliable drag on net returns.