Market definition is common in merger analysis, and often the decisive factor in antitrust cases. This has been particularly relevant in the hospital industry, where many merger challenges have been denied due to disagreements over geographic market definition. We compare geographic markets produced using frequently employed ad hoc methodologies to structural methods that directly apply the ‘SSNIP test’ to California hospitals. Our results suggest that markets produced using previous methods overstate hospital demand elasticities by a factor of 2.4 to 3.4 and were likely a contributing factor to the permissive legal environment for hospital mergers.