Quantitative wildfire risk assessments increasingly are used to prioritize areas for investments in wildfire risk mitigation actions. However, current assessments of wildfire risk derived from fire models built primarily on biophysical data do not account for socioeconomic contexts that influence community vulnerability to wildfire. Research indicates that despite accounting for only a small proportion of high wildfire hazard areas, communities with fewer socioeconomic resources to devote to wildfire prevention and response may experience outsized exposure and impacts. We examined the distribution of simulated wildfire risk versus observed wildfire experience relative to social vulnerability across communities in the Pacific Northwest region of the United States. Using three decades of wildfire occurrence data, we investigated whether socially vulnerable communities were more likely to experience ignitions, fires that escaped initial attack (hereafter 'escaped fires') (>121 hectares), and large fires (>404 hectares), reasoning that each may reveal key insights into the effectiveness of current wildfire risk mitigation and response efforts. We found that communities located in areas with higher wildfire risk or hazard tended to have lower social vulnerability, but that across landscapes east of the Cascade Range, communities with higher social vulnerability were more likely to be exposed to ignitions, escaped fires, and large fires. Our results draw into question whether the current reliance on biophysical data in wildfire risk assessments, absent consideration of community socioeconomic conditions, may perpetuate social inequities by leading to over-investment in well-resourced communities and under-investment in socially vulnerable communities subject to disproportionate wildfire exposure.