We ask whether retail investor attention forecasts stock returns or only trading in an emerging, retail-dominated market. Using weekly Google search volume for 65 heavily-searched IDX80 stocks on the Indonesia Stock Exchange over 2021–2025, we build an abnormal-search measure and estimate predictive panel regressions with firm and week fixed effects and clustered standard errors. Attention is associated with next-week abnormal trading volume: the relation holds across ten specifications, passes a within-week permutation test and a multiple-testing correction, and survives controlling for the contemporaneous price shock, though it shares variance with same-week trading. Attention does not forecast directional returns, and the confidence interval is tight enough to exclude a developed-market-sized price continuation. We find no robust reversal and no concentration in small stocks; if anything it is stronger among large, recognizable names. We also document the keyword-disambiguation problem that non-English tickers create for search-based attention measures.
Copyrights © 2026