This was a longitudinal study utilising the Irish Longitudinal Study on Ageing (n 3849 aged ≥ 50 years) and investigated the relationship between blood plasma folate and B12 levels at baseline (wave 1) and incident depressive symptoms at 2 and 4 years (waves 2 and 3). A score ≥ 9 on the Center for Epidemiological Studies Depression Scale-8 at wave 2 or 3 was indicative of incident depressive symptoms. B12 status profiles (pmol/l) were defined as < 185, deficient low; 185 to < 258, low normal; > 258–601, normal and > 601 high. Folate status profiles (nmol/l) were defined as ≤ 10·0, deficient low; > 10–23·0, low normal; > 23·0–45·0, normal; >45·0, high. Logistic regression models were used to analyse the longitudinal associations. Both B12 and folate plasma concentrations were lower in the group with incident depressive symptoms v. non-depressed (folate: 21·4 v. 25·1 nmol/l; P = 0·0003; B12:315·7 v. 335·9 pmol/l; P = 0·0148). Regression models demonstrated that participants with deficient-low B12 status at baseline had a significantly higher likelihood of incident depression 4 years later (OR 1·51, 95 % CI 1·01, 2·27, P = 0·043). This finding remained robust after controlling for relevant covariates. No associations of folate status with incident depression were observed. Older adults with deficient-low B12 status had a 51 % increased likelihood of developing depressive symptoms over 4 years. The findings highlight the need to further explore the low-cost benefits of optimising vitamin B12 status for depression in older adults.