A feature of recent policy discussion both in Australia and overseas has been a heightened interest in energy taxes and fuel taxes of various kinds. These taxes have been advocated on various grounds, notably their role in discouraging greenhouse gas emissions. At the same time, at least in Australia, greenhouse policy discussion has been redirected more towards small-scale sector-specific interventions, and away from economy-wide measures such as a carbon tax. In this context it becomes of interest to ask, how effective might an energy tax be in reducing carbon emissions? Here we use the term energy tax to mean fossil fuel taxes excluding carbon taxes. A carbon tax is levied on carbon dioxide emissions or some closely related basis, while an energy tax is levied on some other basis such as energy content. The paper presents simulation results designed to address these questions. The simulations are performed using the ORANI model of the Australian economy, in a version containing several energy-specific enhancements. These include greater detail on energy production and use in the database, and a wider range of substitution possibilities in energy production and use in the theoretical structure. The database enhancements include extensive disaggregation of the two largest parts of the energy sector, fossil fuels and electricity. The theoretical developments cover substitution between energy and capital, between different sources of energy, between different techniques of generating electricity, and between different modes of transport. We fmd that a broad-based energy tax would be comparable in effectiveness to a carbon tax in reducing greenhouse gas emissions. This is because like the carbon tax it would bear heavily on the cheaper fossil fuels and would induce emission abatement through fuel switching. Taxes such as a petroleum products tax which excluded the cheaper fossil fuels would be much less effective.