Driving Costs and the Demand for Cars

Gasoline is an important complement to cars.  Other things equal, as gasoline prices rise, miles driven and car sales decline, but more can be learned by separating cars into makes and models, making one a substitute for another.  But the strength of that relationship depends on a number of characteristics.  One make and model, a Honda Civic, may be a strong substitute for another, a Toyota Corolla, and a weak substitute for another, a Chevy Tahoe. 

With the price of related goods for the demand car model in mind, I constructed a table based on the per mile total cost to operate the seven different models over five years assuming the cars were driven 15,000 miles per year.  All the costs except fuel were taken from Yahoo Autos.  Fuel costs were estimated using the average of EPA City and Highway estimates at gasoline price beginning at $2.75/gallon, increasing at $.25 increments to $4.25/gallon.  Fuel economy ranged from 18 City/27 Highway for the Ford Taurus to 51 City/48 Highway for the Toyota Prius. 

This is a simple model; the only cost that changes as gasoline prices rise is fuel.  I expect that other elements of cost would change as the price as well, For example, the price of fuel efficient cars would rise relative to less efficient automobiles both in the new and used markets.  Because I don’t have data to predict the changes in new car purchase price and the used car sales price, I did not modify Yahoo’s cost estimates.  The graph of the per mile total cost of the seven vehicles yielded some expected results and some surprises.  As expected, the total cost per mile of the hybrids rose more slowly than non-hybrids, but not dramatically so.  I anticipated that the total cost per mile of the non-hybrids would be lower at low gasoline prices but would be higher than the hybrids at high gasoline prices.  The Civic and the Fusion both came in non-hybrid and hybrid models and in both cases, the hybrids had lower total cost per mile than their counterparts at $2.75, and that cost differential grew as prices rose.

The reason that gasoline prices did not influence total cost per mile is that they are a relatively small part of overall cost.  Fuel cost never rose above 33% of total cost (Mazda 3 Sports i), and were as low as 17% of total cost at $4.25/gallon (Toyota Prius).  In general, models that were relatively less expensive at low gasoline prices were less expensive at high gasoline prices.

Consumers will try to lower the cost of owning a car, but how?  A more sophisticated inquiry might study the complementarity of makes and models.  Is a minivan a stronger or weaker complement to a large SUV than a crossover?  Likewise, is a compact hybrid a stronger or weaker complement to a compact than a subcompact?