The present study examined the moderating effect of supervisory support on the relationship between self-ratings of self-efficacy and two measures of work performance, namely supervisor-rated performance and a work-output measure. Research was conducted within the South African life insurance industry on a sample of broker consultants (sales promotional staff) ( n = 70 subjects) over a three-month period. Past research found a positive relationship between high self-efficacy and work performance. However, such research did not take account of the influence of leadership factors. It was hypothesised that effective supervisory support would facilitate the attainment of higher levels of performance. Therefore, it was hypothesised that supervisory support should moderate the self-efficacy work performance relationship, particularly in the case of low self-efficacy. As hypothesised, it was found that supervisory support moderated the relationship between self-efficacy and supervisor-rated performance. Performance was highest when low levels of self-efficacy occurred together with high supervisory support. However, supervisory support did not moderate the relationship between self-efficacy and the work-output measure of performance. Limitations of the study, as well as theoretical and practical implications are considered.