High intake of sweet foods such as cakes, cookies, chocolate and ice cream has been reported to be associated with depressive symptoms. However, prospective studies are scarce and no study has been conducted in Asian populations. We prospectively investigated the association between confectionery intake and depressive symptoms in a Japanese working population. Participants were 911 workers (812 men and 99 women; aged 19–68 years) without depressive symptoms at baseline who completed a 3-year follow-up survey. Dietary intake was assessed using a validated self-administered diet history questionnaire. Depressive symptoms were assessed using the Center for Epidemiologic Studies Depression (CES-D) scale. Multiple logistic regression was used to estimate the OR of depressive symptoms according to tertile of confectionery intake with adjustment for covariates. At the time of the 3-year follow-up survey, 153 (16·8 %) workers were newly identified as having depressive symptoms (CES-D score ≥ 16). Confectionery intake was significantly associated with increased odds of developing depressive symptoms. The multivariable-adjusted OR of depressive symptoms for the highest v. lowest tertile of confectionery intake was 1·72 (95 % CI 1·03, 2·86) after adjustment for covariates including dietary factors such as folate, vitamin B6, vitamin B12, n-3 PUFA, Mg, Zn and soft drink (Pfor trend = 0·012). Our results suggest that confectionery intake is associated with a higher risk of depressive symptoms in a Japanese working population predominantly comprised of men.