We exploit an exogenous shock to dark trading volume to identify the effect of dark trading on market quality. Following a 34% reduction in trading on dark venues, we find no evidence that the cost of trade (e.g., effective spreads, realized spreads, price impact, and quoted spreads) changes in a statistically or economically meaningful manner. While our findings stand in contrast to those of several prior studies, supplemental tests confirm that contradictory inferences cannot be attributed to either low power or different stock samples or time periods. Instead, we argue that identification is a key driver in conflicting results. Our research highlights the benefit of structured experimentation from the Securities and Exchange Commission (SEC) for understanding causal effects in capital markets.