With almost 23,000 deals already processed and tens of thousands more in the pipeline, it's possible the $1 billion allocated for the program might have already run out.
The back of my napkin looks like this: Assuming 30,000 deals at $5,0000 a pop and you have spent $1.5 million in a week. Assume that rate constant and you have 6 and a half weeks or so until you exhaust the $1 Billion allocated. I would suspect, in reality, the rate would decay and the program could safely stretch for two months. Unless, of course, administrative costs dramatically exceed the 25% I factored in on top of the $3.5-$4.5K received by consumers, or the money allocated is going elsewhere.
Or, and this may be most likely, the administration is pursuing a cabbage-patch strategy to generate positive -- for both program and government -- publicity.
The other back of my napkin looks like this: There were 1M vehicles sold in June. Sales for July are expected to be up 10% meaning an extra 100K. Less than 3% of total auto-buyers in July took advantage of the program. (Even though the program was not available till the end of July, it was passed end of June, so rational buyers interested in the program would have delayed their purchases until it began.) My initial gut was that this number was low until my wife pointed out that $3500 is not a ton for a used car in reasonable shape.
What this, to me, means is that a good chunk of Cash For Clunkers participants would likely have a bought a new car this year in any event. If 2/3s of the participants would not have otherwise purchased a car then 70% of the projected July over June increase is due to other factors, 10% represent sales cannibalized from future months and the remaining 20% is what we have to show for a 1.5M - 1B dollars of government spending.