Radiative characteristics (albedo, transmission, absorption, and horizontal transport) of stochastic stratocumulus are studied. The spatial distribution of the liquid water content (LWC) is simulated as a random process with a one-dimensional lognormal distribution and the power-law spectrum (inhomogeneous interval cloud structure). Stochastic cloud top geometry is simulated as a random Gaussian process with exponential correlation function. It is shown that the stochastic geometry and the inhomogeneous internal cloud structure affect nearly identically the mean and the variance of albedo and transmission. The contribution from the stochastic geometry to the variance of horizontal transport predominates. Fluctuations of cloud top altitude decrease the accuracy of cloud absorption retrieval by an order of magnitude. Simultaneous measurements of fluxes in the visible and near-IR spectral ranges can be used to investigate small-scale (~0.4 km) variations of absorption.