Major depressive disorder (depression) is a complex condition that involves multiple physiological mechanisms, spanning a range of spatial scales. Altered cortical inhibition is associated with treatment-resistant depression, and reduced dendritic inhibition by somatostatin-expressing (SST) interneurons has been strongly implicated in this aspect of the pathology. However, whether the effects of reduced SST inhibition on microcircuit activity have signatures detectible in electroencephalography (EEG) signals remains unknown. We used detailed models of human cortical layer 2/3 microcircuits with normal or reduced SST inhibition to simulate resting-state activity together with EEG signals in health and depression. We first show that the healthy microcircuit models exhibit emergent key features of resting-state EEG. We then simulated EEG from depression microcircuits and found a significant power increase in theta, alpha and low beta frequencies (4 - 15 Hz). Following spectral decomposition, we show that the power increase involved a combination of aperiodic broadband component, and a periodic theta and low beta components. Neuronal spiking showed a spike preference for the phase preceding the EEG trough, which did not differ between conditions. Our study thus used detailed computational models to identify EEG biomarkers of reduced SST inhibition in human cortical microcircuits in depression, which may serve to improve the diagnosis and stratification of depression subtypes, and in monitoring the effects of pharmacological modulation of inhibition for treating depression.