Depression is associated with impairment of cognitive functions, and especially executive functions (EFs). Despite the fact that most depressed patients experience recurrence of episodes, the pattern and the severity of executive impairment have not been well characterized in this group of depressed patients. We asked if and to what extent these patients were impaired on a range of neuropsychological tests measuring EFs, and also when confounding factors were adjusted for. Forty-five patients (aged 19-51 years) with moderate to severe (Hamilton score >18) recurrent major depressive disorder (DSM-IV) were compared to 50 healthy controls matched on age, education, gender and intellectual abilities. The subjects were administered a set of neuropsychological tests that assesses sub-components of EFs. The depressed patients were impaired compared to the control group on all selected tests, with a severity of impairment within -1 standard deviation from the control group mean. The group difference was statistically significant for eight of the 10 EFs that were assessed. These were measures of verbal fluency, inhibition, working memory, set-maintenance and set-shifting. The group difference was still significant for all sub-components except for set-shifting (Wisconsin Card Sorting Test) and planning (Tower of London), when additional medication and retarded psychomotor speed was adjusted for. In conclusion, the depressed subjects were mildly impaired across a wide range of EFs. This may have a negative impact on everyday functioning for this group of patients.