Purpose Working alliance is one of the most important common factors for successful counseling/psychotherapy outcomes. Based on the empirical literature about working alliance, it seems that self-determination and self-efficacy theory (SDT/SET) can potentially be used as a motivational model to explain the relationship between working alliance and vocational rehabilitation (VR) outcomes. The purpose of this study is to evaluate three primary SDT/SET constructs, autonomous motivation, expectancy and engagement, as mediators for the relationship between working alliance and stages of change (SOC) for employment. Methods A serial multiple mediation analysis (SMMA) was computed to evaluate autonomy, outcome expectancy, and VR engagement as mediators of the relationship between working alliance and SOC for employment in a sample of 277 people with chronic illness and disability (CID) receiving services from state VR agencies in the United States. Results The SMMA results indicated that working alliance was positively associated with SOC for employment (total effect), while the direct effect between working alliance and SOC for employment was not significant after controlling for the effects of the mediators, indicating significant mediation effects. The mediation effects were estimates of the indirect effects for working alliance on SOC for employment through (a) autonomous motivation, (b) outcome expectancy, (c) VR engagement, and (d) autonomous motivation, outcome expectancy and VR engagement together. Conclusions The results indicated that a strong working alliance has the benefit of helping consumers develop autonomous motivation to work and increasing their vocational outcome expectancy and engagement in VR services, leading to employment.