This paper reveals significant errors in a key variable in international trade: tariff rates. The issues arise from incomplete reporting, leading to measurement error from false interpolation by the data provider, the World Integrated Trade Solution (WITS), and selection bias from dropping tariffs when no trade is recorded. I develop a novel interpolation algorithm to correct these issues and construct a global tariff dataset. Reestimating recent studies relying on WITS data highlights the importance of correcting these errors. Studies using cross-country tariff variation, such as estimates of the trade elasticity, are particularly sensitive to the mishandling of the tariff data.