Some empirical relationships among car repair cost, engine size, and production year. April 10, 2006. 

Disclaimer - Data shown in this page are an amateur curiosity only. 

What is this page for 

 I collect here some automotive data - about cars intended for people private transportation. The dataset contains most of mass production cars (for mass production I mean not coming from small firms who produce just few samples; I don't mean accessible purchase costs). I have a few curiosity that I try to satisfy by simple analysis.  The questions arise for a simple reason, because I want to change my car with a new or a nearly new one and want to evaluate costs, performances, and safety of the different car models.

First curiosity: are bigger engine cars most expensive  to repair after an accident occurs?

 The possible intuitive answer has to end points. A cheaper and small car could also be possibly cheap to repair - damages cannot be more than the car value (well, this excludes concerns about driver's damage, but this point has to be considered in a separate safety chapter). From the other side, a bigger, better engineered car should be relatively cheap to repair - if it has really well engineered. But what's in the middle? 

Let's look at this graph, showing the repair costs (indexed as an integer number from 1, the minimum, to 20, the maximum) versus the engine volume (in cc, or cubic centimeters, cm^3).

This plot shows that cars whose engines are up to about 1200 cc are cheaper to repair than cars whose engines are bigger than 2500 cc, despite some deviation from the rule. Between 1200 and 2500 cc there is a small trend saying smaller is cheaper, but deviations are huge - you can easily find 2000 cc cars cheaper than many 1500 cc cars.  So, smart buying is possible!

The plot also shows out of production cars. Apparently it seems that more modern small cars  are cheaper to repair than out of production cars, whereas for engines greater than 3000 cc the trend is the opposite - out of production car used to cost less! I can only speculate about the underlying reasons for this, but I am not really interested now.

Second curiosity: are newer models cheaper to run than older ones?

Here the questions seems to have an intuitive answer - more modern projects should be take advantage of newer engineering techniques. But, as we see the next plot, the tendency to decrease repair price seems not to decrease anymore since 1998 - now I wonder why

Anyway, from the plot (larger version) you can see that the repair cost does not decrease anymore.


For now it is enough. More analysis may follow in the future. Anyway, if you have questions, drop me a line.

Ciao for now! 

Go back to my home page