Understanding the feasibility and cost of adaptation is essential to management of the global climate. Unfortunately, we lack general estimates of adaptive responses to almost all climatological processes. To address this for one phenomenon, we estimate the extent of adaption to tropical cyclones (TCs) using the global cross-section of countries. We reconstruct every TC observed during 1950-2008 to parameterize countries’ TC climate and year-to-year TC exposure. We then look for evidence of adaptation by comparing deaths and damages from physically similar TC events across countries with different TC climatologies. We find that countries with more intense TC climates suffer lower marginal losses from an actual TC event, indicating that adaption to this climatological risk occurs but it is costly. Overall, there is strong evidence that it is both feasible and cost-effective for countries with intense TC climatologies to invest heavily in adaptation. However, marginal changes from countries’ current TC climates generate persistent losses, of which only ~3% is “adapted away” in the long run. These findings are consistent with the Envelope Theorem (Nordhaus, 2010).