Tag Archives: fuel

Buying a fuel-efficient used car versus a new Prius

Update: I just found this article, which has the same point as mine, and assumes domestic production. It also pre-dates mine – that will teach me not to use google!

As one who drives a 2000 Honda Civic, which in many real-world driving tests gets almost exactly 30 MPG consistently (I keep pump logs), I wonder if I would actually create a net reduction in gas consumption by buying a Prius or other car. Now, I’m assuming a new Prius, but buying a used one is perfectly valid, and is outside the scope of this article. If you want to do that, more power to you – that’s a good choice hands down. Yet, most people I know would opt for a new one. Based on various seemingly valid estimates, it takes 113,322,000 BTUs to create and import a brand new Prius. It takes 0 BTUs to park a used Honda Civic on a lot until someone buys it.

So, for the data used to obtain this, I looked all over for real road tests of the Prius fuel economy.  This one seemed to fit well with everything else I’d been reading, giving an average range of 42.6-45.2 MPG.  So, I’m going to say 43 MPG.

The Prius uses enough gas to create 2639.53 BTUs per mile.  The Civic: 3783.33.  At these rates, and coupled with the initial component of the BTUs used to manufacture and import the Prius, we come up with the following:

$latex 2639.53x + 113322000 = 3783.33x$

$latex x = 99075.01$

Graphing this in gnuplot, we get the following:

plotSo, one would have to drive almost 100,000 miles to get an advantage over simply buying a used Civic.  Interesting.  Here’s the GNUPlot Plot File for anyone that’s interested.