Background: School systems have struggled to staff special education positions for decades (Billingsley, 1993; Cowan et al., 2016; Mason-Williams et al., 2020), but recent data suggest special education staffing challenges are now a crisis. As of October 2023, 21% of schools, and 30% of high-poverty schools, reported at least one special educator vacancy (School Pulse Survey). The staffing problems disproportionately affecting students with disabilities is an equity issue; student with disabilities cannot receive the services they need to succeed in school without qualified special educators. In this policy context, several states and districts are offering financial incentives to special educators with the goal of addressing special educator shortages (Putnam & Gerber, 2022). One recent example is from Hawai'i Public Schools, which implemented a bonus policy starting in fall 2020 that raised the salaries of all special educators in the state by $10,000 (McCoy, 2022). This special education bonus, combined with additional bonuses for teachers working in historically hard-to-staff schools, resulted in special educators in the hardest-to-staff schools making up to $18,000 more per year than prior to the bonus policy. Although prior evidence on teacher financial incentives suggest bonuses may address long-standing special educator shortages in the state, there is no existing causal evidence on the impact of bonuses specifically offered to special education teachers on these critical staffing challenges (Gilmour et al., 2024). Research Questions: We investigated three research questions--RQ1: What was the average causal effect of the $10,000 bonus policy on special education teacher shortages in Hawai'i, as measured by the proportion of positions that were vacant and/or by the proportion of positions filled by unlicensed teachers? RQ2: To what extent were these effects driven by changes in patterns of teacher attrition and movement between special education and general education teaching positions? RQ3: How did the effect of the $10,000 bonus policy vary between hard-to-staff schools in which all teachers received additional bonuses and other schools? Setting, Data Sources, and Participants: We used longitudinal position-level staffing data from Hawai'i Public Schools from the 2014-15 through the 2022-23 school years. The Hawai'i data are unique in that they provide an annual snapshot from October 1 of each school year of every school staff position in the state, whether or not the position was filled, along with indicators for whether positions that are filled are filled by a teacher who meets the state's licensure requirements. This allowed us to create two position-level measures of staffing challenges (positions that were entirely vacant and/or positions that were filled by unlicensed teachers) in addition to annual indicators for the attrition and hiring of individual teachers in the state. The analytic sample included seven districts, 15 complex areas, and 261 schools with a sample size of more than 115,000 positions across 9 school years. Prior to the bonus policy 5-6% of special education positions were vacant compared to 1-3% of not special education positions. Research Design and Analysis: We investigated RQ1 and RQ2 by estimating difference-in-differences (DID) models that exploited the fact that an untreated group of teachers (general education teachers) were ineligible for the $10,000 special education teacher bonus. For RQ3, we estimated triple difference (DDD) models to explore heterogeneity in the impacts of the bonus policy across different "tiers" of schools that were identified for additional schoolwide bonuses under the state's bonus policy. Results: For RQ1, our DID estimates suggested that the introduction of the special education teacher bonus policy reduced the proportion of vacant special education teaching positions by 32%, or 1.2 percentage points, and the proportion of special education positions that were vacant or filled by an unlicensed teacher by 35%, or 4.0 percentage points. These findings were robust to a variety of model specifications. We therefore interpreted these estimates as the average causal effects of the special education bonus policy on these proxies for special education teacher shortages, and we concluded that the $10,000 bonus was sufficiently large to significantly address special education teacher shortages in the state. When we explored potential mechanisms for these effects as part of RQ2, we found that the bonus policy had no significant impact on special education teacher retention. Instead, we documented a large increase in the proportion of open special education positions that were filled by general education teachers in the state after the introduction of the policy. Finally, for RQ3 we explored heterogeneity in the impacts of the bonus policy across different "tiers" of schools that were identified for additional schoolwide bonuses under the state's bonus policy due to a variety of factors including prior vacancy rates. For example, all teachers in Tier 4 schools received bonuses of $8,000 under the policy; special education teachers in Tier 4 schools received $18,000 bonuses, whereas general education teachers in the same school received $8,000 bonuses. Estimates from the DDD models suggested that the impact of the bonus policy on special education vacancies was driven almost entirely by impacts in Tier 4 schools, whereas the impact of the bonus policy on positions that were vacant or filled by unlicensed teachers was significant across the different school tiers but larger in Tier 3 and Tier 4 schools. Conclusions: Increased special education shortages threaten the ability of schools to provide services to students with disabilities. In this study, we found that bonuses attenuated staffing problems by attracting general education teachers with special education certification to fill special education positions. The bonus policy was differentially effective at addressing special education teacher shortages in historically hard-to-staff schools. Interestingly, the bonus did not improve special education teacher retention, suggesting the need for more policies and practices to address other variables associated with special education attrition, such as working conditions. Future work will be necessary to determine the long-term impacts of the policy, including whether the bonuses induced more potential teachers to pursue a special education teaching position in the first place.