February 21, 2013

Car Depreciation Over Time

You may have heard someone say that "new cars lose half their value when you drive them off the lot" or something similar.   While its true that cars depreciate a hefty amount in their first year its not that steep of a drop that fast.   After the car is used it tends to depreciate at a more gradual pace year over year.

I pulled sales prices off Edmunds for several car models :
Toyota Camry, Ford Crown Victoria , Honda Accord, Chevy Malibu , Cadillac Escalade
Dodge Grand Caravan , Acura RL
 I chose those models semi-randomly.   I needed to pick some cars that had been on the road over many years.   I also wanted some cars from various manufacturers.

First lets look at the actual dollar value for a few typical used sedans based on model year.

(click image for full size)
As you can see there  the pattern of depreciation is pretty similar.  

Now lets normalize and look at % values from the start over a number of year.   The chart below looks at prices starting in either 2011 or 2012 and then shows the % of value retained each year after that.   So for example the cars are generally down to 50% of the original value by the 5-7 year period and then down to only 20% by 8-13 years.

This is a pretty small sample of cars so I'm not going to try and take much meaning out of this as far as differences between models.    The idea here is just to look at depreciation for cars in general.

Note: This is a crude approximation of depreciation trends.  For one I'm ignoring inflation and not figuring the initial retail cost of the older cars when new versus their current resale value. 



  1. My strategy is to never, ever, drive the car off the lot. My balance sheet looks great, but I am starting to question my financial model; the risk, utility value, and maintenance cost weightings may need tweaking.

  2. can you post the numbers? i'm wondering what the drop is as a percentage of the current value (as opposed to the initial value)

  3. I agree with Fat Bob. Depreciation seems to follow a curve of going down perhaps 20% per year, with a floor (e.g. a toyota that still runs is worth about $1000 no matter how old it is). It seems that a new car doesn't depreciate much faster than a used car at all, relative to the initial cost. It just starts at a lower point on the curve, which does lower the depreciation in absolute dollars.

  4. If you want to see the numbers for each car then you can follow the links for each car to the Edmunds site. IT lists the current used resale value by model for each year. For the chart with the % values I started with new MSRP as 100% then use the used values after that. Now this does make an assumption that future depreciation will follow the trend of previous years, i.e. that deprecation from 2013 to 2015 will be simillar path as 2011 to 2013. Also even within a specific model there are other factors from year to year that might impact depreciation a little such as major model changes that could add or subtrack standard options nad impact the resale. Like say if in 2008 it was standard to have cloth seats then in 2009 they added leather as standard then there would be a difference between 2008 and 2009 cars values based on those leather seats.

    And of course depreciation varies between make/model. Toyotas tend to depreciate slower than Chevys in general. Expensive cars may depreciate faster than less expensive.

    My goal here was to look at the general pattern of depreciation with some examples. Its not exhaustive nor definitive of course as my sample size is fairly low.


  5. Here are the numbers for Camry:

    2013 $22,235
    2012 $18,189
    2011 $14,871
    2010 $13,835
    2009 $12,072
    2008 $10,144
    2007 $8,946
    2006 $7,650
    2005 $6,794
    2004 $6,737
    2003 $5,797
    2002 $5,362
    2001 $4,155
    2000 $4,085
    1999 $3,648
    1998 $3,332
    1997 $3,152
    1996 $2,794
    1995 $2,416
    1994 $2,134
    1993 $1,945
    1992 $1,754
    1991 $1,559
    1990 $1,301

    That equates to year-to-year changes of :

    2012 -18%
    2011 -18%
    2010 -7%
    2009 -13%
    2008 -16%
    2007 -12%
    2006 -14%
    2005 -11%
    2004 -1%
    2003 -14%
    2002 -8%
    2001 -23%
    2000 -2%
    1999 -11%
    1998 -9%
    1997 -5%
    1996 -11%
    1995 -14%
    1994 -12%
    1993 -9%
    1992 -10%
    1991 -11%
    1990 -17%

  6. I should have probably titled this one 'Some examples of Car Depreciation over Time'.


  7. With a chart like that? You've got to be thinking calculator as a follow up. Am I right?

  8. PK, sounds like a good idea. Let me know when you've finished it. ;)

  9. Haha, can't tell if you're joking so I'll make this offer: send me the data for the last chart and I'll do one.

  10. Hallo. Your charts are very good. I was the writer of the article in wikipedia for "car costs". Can I use your charts for wikipedia, with Creative Commons license, citing obviously the author? Thank you

    1. Yes you may use my charts. However you should probably also cite Edmunds as a source.


    2. Thank you so very much, You'll need to give consent by writing it, otherwise the chart in wikipedia will be deleted.

      Can you just copy-paste this text

      and send it to:

      your chart is here (you'll need this on the text):

      thank you so much again

  11. Sorry for commenting on a long dead post, but I wonder if simply looking at prices for a given make/model is the right way to go. For example, the 2010 Camry is a much different car than the 1990 Camry. I would be more interested in how a given edition of a make/model performs over time. So, say the Camry was updated in 2010 and again in 2014. How do the 2010-2013 Camry's do year after year.

    Also, plotting the change in $ or % value with time vs time would be useful. The slope of $ vs time is negative, but the magnitude decreases with time. It may be easier to see where the slope begins to level out. I would argue that the age of the vehicle you select should be dictated by the point where the slope of the depreciation curve levels out.


Blog Widget by LinkWithin