Respiratory system reactance (Xrs) measured by the forced oscillation technique (FOT) is theoretically and experimentally related to lung volume. In chronic obstructive pulmonary disease (COPD), the absolute volume measured by body plethysmography includes a proportion that is inaccessible to pressure oscillations applied via the mouth, that is, a “noncommunicating” lung volume. We hypothesized that in COPD the presence of noncommunicating lung would disrupt the expected Xrs-volume relationship compared with plethysmographic functional residual capacity (FRC_pleth). Instead, Xrs would relate to estimates of communicating volume, namely, expiratory reserve volume (ERV) and single-breath alveolar volume (Va_SB). We examined FOT and lung function data from people with COPD ( n = 51) and from healthy volunteers ( n = 40). In healthy volunteers, we observed an expected inverse relationship between reactance at 5 Hz (X5) and FRC_pleth. In contrast, there was no such relationship between X5 and FRC_pleth in COPD subjects. However, there was an inverse relationship between X5 and both ERV and Va_SB. Hence the theoretical Xrs-volume relationship is present in COPD but only when considering the communicating volume rather than the absolute lung volume. These findings confirm the role of reduced communicating lung volume as an important determinant of Xrs and therefore advance our understanding and interpretation of FOT measurements in COPD. NEW & NOTEWORTHY To investigate the determinants of respiratory system reactance (Xrs) measured by the forced oscillation technique (FOT) in chronic obstructive pulmonary disease (COPD), we examine the relationship between Xrs and lung volume. We show that Xrs does not relate to absolute lung volume (functional residual capacity) in COPD but instead relates only to the volume of lung in communication with the airway opening. This communicating volume may therefore be fundamental to our interpretation of FOT measurements in COPD and other pulmonary diseases.