The cerebral cortex is known to follow a hierarchical organization, which supports information processing and ultimately determines behavior. The hierarchy has been observed in brain connectivity and cytoarchitecture, as well as the timescale of spiking activity. However, whether further dynamical features of cortical activity follow a hierarchical organization remains an open question.To address this question, we measured single-neuron statistics from the resting-state activity of macaque neocortex (N=6) from several cortical areas (V1, V4, DP, 7A, M1, PMd, vlPFC). We found some non-systematic differences when considering univariate spiking statistics, such as firing rate, variability, and correlations. When considering the high dimensional multivariate distribution of the same spiking statistics, significant and systematic differences between all cortical areas were revealed, consistently across different subjects and laboratories. Furthermore, we introduce a new dynamical distance, defined as the higher-dimensional Wasserstein distance between multivariate spiking statistics of cortical areas. We found that the dynamical distance strongly correlates with an anatomical definition of the cortical hierarchy (Figure 1). Therefore, we demonstrate a strong link between brain structure and function. Our results suggest that the dynamical distance could be used to help understand information processing in the brain, enabling comparisons between different experimental setups and subjects. Additionally, the multivariate spiking statistics uniquely characterize cortical areas, which could be exploited for parameter inference in brain models.Figure 1: Dynamical distance is the Wasserstein distance between multivariate spiking statistics of each area. Hierarchical distance is the log ratio between neuron densities. Strong correlation is observed, outlier excluded.