This paper proposes a means of using stochastic diffusion processes to model the total consumption of electrical power (including distribution and transport losses) in Morocco, as recorded by the official data for total sales published by Office Nationale de I'Electricite (ONE), the Moroccan electricity authority. Two models of univariate stochastic diffusion were used: the time-homogeneous Gompertz Diffusion Process (HGDP) and the time-non-homogeneous Gompertz Diffusion Process (NHGDP). The methodology proposed is based on the analysis of the trend function; this requires the analyst to obtain fits and forecasts for the consumption of electrical power by means of the estimated trend function (conditioned and non-conditioned). This latter function is obtained from the mean value of the process and the maximum likelihood estimators (MLE) of the parameters of the model. This estimation and the subsequent statistical inference are based on the discretised observation of the variable "electricity consumption in Morocco", using annual data for the period 1980-2001. The fit and forecast are improved by using macroeconomic exogenous factors such as the gross domestic product per inhabitant (GDP/inhab), the final domestic consumption (FDC) and the gross fixed capital formation (GFCF). The results obtained show that NHGDP, (with the above three exogenous factors) provides an adequate fit and medium-term forecast of electricity consumption in Morocco. (c) 2005 Elsevier Ltd. All rights reserved.