If vehicle sales are projected to triple by 2050, how come cutting fuel use per kilometre by half doesn’t result in a 1.5 times increase in fuel use?

This is due to a number of factors (based on details of the IEA ETP 2008 projections):  first, we do not quite expect a tripling of car stocks, but its close (around 2.8x in our mid-range projection).  Second, we expect a slight reduction in travel per vehicle over time as more vehicles become available around the world – so triple the number of cars does not mean triple the total kilometres travelled.  Third, as we say in our document, we assume a few “flanking measures” to occur, such as a reduction in the “gap” between tested efficiency and actual in-use fuel economy – so it’s not triple the volume of fuel used either. Overall therefore, if we cut fuel use per kilometre in half, we can actually save over 6 billion barrels of oil per year by 2050 – even with a growing global fleet.