This paper provides new insights into the longstanding empirical issue of whether the type of workplace saving plan (a "traditional" registered pension plan or RPP, a "flexible" group registered retirement savings plan or group RRSP, and a "hybrid" arrangement of the two) affects employee voluntary job separations. We use a Canadian employer-employee matched dataset that provides information on both job transitions and the types of workplace saving plans being held by employees and offered by employers. This dataset allows us to control for employee self-selection and firm fixed effects. The standard prediction from implicit contract theory suggests that traditional pensions reduce quit rates but flexible plans have little effect due to their portability. The results are partially consistent with this prediction. Implications of these findings for current public policy are discussed.