Do not put your faith in what statistics say until you have carefully considered what they do not say.
-William W. Watt
Mr. Watt gives us good advice and I have looked at what my few small charts show and I do understand what they do not show. To really get into the connections between the number of Lemon Law cases and the number of vehicles sold by each manufacturer in a given year might make an interesting book. One could connect manufacturing technology, quality of process, quality of materials, the attitude of manufacturers and their dealers to consumers, unions, the effect of the quarterly report and how it drives sales, the legal ramifications of building and selling defective vehicles and on and on.
Statistics are interesting provided they show something startling or prove something your intuition suggested was a fact, but that hadn’t been proved. For example: I know, that’s know not think, that as a country and as individuals we are prisoners of our national business model. As prisoners, we are punished over and over and we feel it in that place of enduring pain, our wallets.
After spending some years teaching TQM [Total Quality Management-Deming] and SPC [Statistical Process Control] I came to the understanding that manufacturing high quality components [there are on the order of 15,000 components in the average modern automobile] is subject to considerable variation, even in the absence of the human element. That variation in the manufacture of components, the variation in the processes of assembly of those components, the variation in materials practically guarantees that a surprising percentage of all vehicles manufactured are going to result in the assembly of a Lemon Automobile. By Lemon I mean a vehicle that meets the legal tests of the various state’s Lemon Laws. The question then becomes, why does one manufacturer who produces a million vehicles produce 5% lemons and another manufacturer who also makes a million vehicles, produce less than 1% lemon vehicles. That’s an extraordinary difference. That means the 1st manufacturer made 50,000 lemons and the other made less than 10,000 lemon vehicles.
I think I know the answer and I am betting those of you who read this also have good ideas of your own. What I wanted to do here is show, very directly, that there is a correlation between manufacturing skills, manufacturing quality and the number of defective vehicles-lemons-produced by specific manufacturers. It is, I think, a good idea to temper statistics with common sense and the skilled use of the best computer available-the one between your ears.
Statistics are no substitute for judgment
-Henry Clay American Statesman, 1777-1852
Figure 1 : Car & Light Truck Sales in the US for 2007
As you have noted these are sales, not production numbers. When one looks at production it will be seen that GM and Toyota are very close in total production. The reason we include production numbers is that showing the percentage of lemon law cases by manufacturer we had during 2007 is interesting but not nearly as telling as when these percentages are compared to numbers of specific makes produced.
Look at Figure 2. Assume for a moment that the total number of lemon law cases for 2007 was 1,000. Now look at the percentages for GM and Toyota. For roughly the same number of vehicles produced, there were 166 GM cases and 35 Toyota lemon law cases. Think about this in terms of money, dollars and cents. Remember, GM is in so much trouble it is closing plants and making serious efforts to get costs under control. Toyota is not closing plants, they are in fact, steadily increasing production and market share. As an American this &^^&%& me off.
Figure 2 : Percentages of Lemon Law Cases by Vehicle Manufacturer 2007
Let’s have one more comparison to really make the point. Look at Mercedes. Their percentage of our total lemon law cases last year was 14%. That would be 140 cases based on my proposed 1,000 total cases. Now, let’s pop back up to Figure 1. We note that their total sales were 226,003 vehicles. I got the production numbers from a reliable source on the Internet.
The higher the percentage of lemon law cases for the smaller number of vehicles sold is the key point of these statistics. It is an indication that something is seriously wrong. It would be well to remember that we are but one Lemon Law firm among many all across the country. Country wide I am sure that there are differences in percentages by make. For example, while I do not know this as a fact, I would guess that more pickup trucks are sold in Alabama than Mercedes Benz. Buying habits differ. However, I doubt there would be much difference if these case statistics were from a sample that included all the lemon law firms in the US. Therefore, while only two basic statistics were used, they are in fact very revealing. I leave it to you, the reader to think about the possibilities.
Because I am an engineering-quality person by training, one key thing repeats over and over to me.
It costs a lot to make bad products.