Any model used to derive projections of future climate or assess its impact constitutes a particular simplification of reality. To date, no model building process can guarantee full “objectivity” in the choice of model assumptions and parameterization. In this connection, researchers have introduced a number of stylized integrated assessment models, which attempt to represent the full time-dynamic non-linear causal loop between accumulated emissions, economy and climate, yet in a aggregated, simplified fashion to enable extensive uncertainty analysis with respect to both structural and parametric uncertainty. In this work, we put forward a simplified system dynamics integrated assessment model which simulates the global economic growth, corresponding emissions, global warming and caused by its secondary effects economic losses. While generally our model follows the same logic as DICE and other models of this kind, it pays more attention to the mechanism of the emission reduction. Mitigation is assumed to be done through the allocation of a certain fraction of the total output into enhancing carbon and energy efficiency. The model enables exploring effects of mitigation scenarios defined via carbon tax. We explore the structural sensitivity by examining five alternative climate sensitivity functions and use the "mutual compatibility integration" approach to synthesize the information from the five alternative model versions.