The advent of the fifth generation of wireless networks forecasts an incredible increase in throughput, number of simultaneous connections and low latency to provide the full set of capabilities. But this unprecedented increase in capacity must not lead to an energy crunch. Therefore, 5G requirements specify it must be achieved at a similar or lower power consumption as todays networks. To evaluate the performance of large representative cellular networks and to capture the major factors involved in the energy consumption process, representative and accurate models are needed. In this paper, we develop a tractable and efficient model based on spatial fluid modeling which reduces the analysis complexity. Moreover, the effectiveness and accuracy of the proposed model are shown through a comparison with the results obtained by Monte Carlo simulations. Afterwards, our simulations results point out that energy efficiency is better in a small cell network compared to a macrocell one and is not related to the density of user equipments.