The energy transition from fossil fuels to alternative energy sources has important consequences for technological change and resource extraction. We examine these consequences by incorporating a nonrenewable resource and an alternative energy source in a market economy model of endogenous growth through expanding varieties. During the energy transition, technological progress is nonmonotonic over time: It declines initially, starts increasing when the economy approaches the regime shift, and jumps down once the resource stock is exhausted. A moment of peak-oil does no longer necessarily occur, and simultaneous use of the resource and the alternative energy source will take place if the return to innovation becomes too low. Subsidies to research and development (R&D) and to renewables production speed up the energy transition, whereas a tax on fossil fuels postpones the switch to renewable energy.

This article analyzes an economy where both nonrenewable resources and a costly energy resource are essential inputs in production. The extraction of the nonrenewable resources leads to emissions that increase the probability of a catastrophe. We find that, in contrast to the constant-probability case, the endogenous probability of a catastrophe implies that some nonrenewable resources might optimally be left in the ground. The larger the effect of the fossil energy use on the probability of a catastrophe, the fewer nonrenewable resources should be extracted and the earlier should be the switch to the renewable substitute. The richer a country, the earlier it should shift to the energy substitute. In the trade-off between higher consumption and a higher probability of catastrophe, even small probability changes are likely to be more important for the planner than higher consumption.

In contrast to the standard approach in energy economics, this paper assumes that physical capital accumulation is relatively more energy-intensive than consumption. By means of Gaussian hypergeometric functions, we provide a closed-form representation of the optimal solution paths of our variables in levels, whatever the initial conditions (i.e., global dynamics). We find that, in general, the optimal trajectories are nonmonotonic. P.S. Dasgupta and G.M. Heal [Review of Economic Studies (special number), 3–28 (1974)] pointed out this result (local dynamics) for the optimal consumption in a model with identical technology for both physical capital accumulation and consumption. However, our paper introduces five novelties with respect to their study, namely, global dynamics, the importance of the proportion of nonrenewable energy resources to endowment of physical capital [S(0)/K(0)], the role of technical progress, U-shaped behavior of consumption, and nonmonotonicity of resource extraction.