I just changed cars, and figured out a system (not necessarily original with me) for comparing the value of cars I was considering buying.

THE BACKGROUND FACTS: My Beautiful Wife and I have for years been great fans of the Buick LeSabre. She is currently driving our fourth LeSabre. We wanted to buy a LeSabre to replace our 14 year-old Cadillac DeVille, which had high fuel/high cost fuel usage and also some incipient front end trouble which would cost more to fix than the vehicle was worth. Trouble is, in 1999 or so the LeSabre was changed, downsized, and we instantly rejected the 2003 LeSabre we test-drove. But the logical successor is the Buick Park Avenue. So off to an online car locator source, in this case Autotrader, where I located six Park Avenue possibilities in the Indianapolis area. I ran Carfax reports on all six, and their histories and reported conditions were all satisfactory.

My plan was (and is) to drive the "new" vehicle until 100,000 miles, and then change again.

The candidate cars I found ranged from a 2000 Park Avenue for $8,977 with only 55,000 miles to a 2003 Park Avenue with 61,000 miles, as I recall, for about $13,700 or so. I didn't really care about what the model year was; I was interested in the miles to be driven between the present and the planned 100K trade-in point.   

So I did the subtraction, 100,000 less present miles, and sorted the cars in descending order of the result, what I'll call "bought miles", the measure of the real use I expected to buy. For this purpose I ignored my present trade-in and the eventual trade-in value when I change again at 100K.

Now I divided the dealer's asking price in each case by the respective bought miles, resulting in a cost-per-bought-mile ranging from 21 cents to 39 cents, and ordered the cars in ascending order of cost-per-bought-mile.

I found that the least costly three cars (by asking price) remained the three least expensive, but that the lowest-price vehicle, the 2000, now moved from number-one to number-two position.

So my B.W. and I looked first to a 2002 Park Avenue with 61,500 miles. We were impressed with it, and decided to look no further. The price had just two days before been dropped from about $13,500 to $10,977. Given the car's ranking on the cost-per-bought-mile test and the recent substantial price cutback, I didn't even try to dicker, but went on to consideration of trade-in of the Cadillac.

Kelley Blue Book and Edmunds both estimated the trade-in value of the Cadillac as about $1,250, based on my describing its condition as "fair". I told the salesman that I had trade-in value reports from those two organizations, and waved the printouts, but didn't disclose
the figures. He looked at whatever reference they use and made a quick airplane guess of $1,500! However, after looking at the Cadillac in detail he cited some body damage which he said would cost around $500 to fix (and I entirely agree with him there), and offered me $1,000. That's fine; that's just the number I had told my B.W. beforehand that I would look for and accept. So we had a deal.

So now we're the proud and pleased owners of a one-owner, well-maintained 2002 Buick Park Avenue Ultra (which means supercharged), with all sorts of bells and whistles, such as multi-CD changer, leather seats, built-in garage door opener, sun roof, rear bumper proximity sensors, and on and on. No longer will I have to buy premium gas, as I did with the Cadillac, and the miles per gallon are A LOT better than the Cadillac's mileage, too. I figure the gas cost alone is going to save me about $25 a month.

But, aside from exulting about my new toy, I wanted to share the cost-per-bought-mile approach, which I merely stumbled upon, and which is novel to me.

Dave Oesterreich