Buy Ford, Sell GM, Hold Toyota (but don't take investment advice from me...)

GM and its employees are taking pride in the wrong things, apparently. What about pride in quality products and pride in profitability? The primary goal isn't market share, unless you're so screwed up like GM where you have excess capacity that you'd better fill, or you have to pay employees to not work.

Toyota DOES build product for dealer orders where an end driver/customer hasn't yet ordered it... so saying "never" builds without a human order depends on what your definition of human customer order means.

You're right about Toyota building product to forecast. That is what Ohno said. I think you missed the "not" in:

What Ohno is saying is not that Toyota never builds an automobile before they have a firm sale with a live human customer

Gotcha. Double negatives aren't very lean! :-)

There is still a lot of opportunity for improvement at Toyota. The fact that they do build to dealer orders without an actual customer order alludes to the fact that there is still waste in their system. I personally would like them to remove more waste and lower the prices customers have to pay for their products.

Onosan talked about type I, II, and III companies in the New Production System Book. He was most interested in the ones who did not keep even final product inventory.

Toyota Motor Manufacturing builds cars based on orders from Toyota Motor Sales, which orders cars based on dealer orders, which are based on forcasted customer sales. Toyota is apparently much better at this than GM, because according to the 12/11/2006 inventory data from Automotive News, Toyota had a 41 day inventory vs GM's 91 day supply of cars and light trucks and an industry average of 73 days. Toyota must be making what the customer wants because the cars they build spend less time on dealers' lots. That does not mean that there is not room for improvement, but the only company that had a lower inventory level than Toyota was BMW with a 25 day supply.

Personally I think that GM should not focus on maintaining the #1 rank in vehicles produced. As long as they are losing an average of almost $2,500 per vehicle (according to the 2006 Harbour Report), each additional vehicle made only helps spread the "legacy costs" out (assuming vehicles don't have to be further discounted because supply is higher than demand), and does nothing to actually reduce cost. It is important to remember one of the reasons behind JIT - reducing cost per part has no impact if the parts are not sold at a price that recoups the cost.