But why? If I want a lower price I don’t care where they have to shave it from to do that do I? I don’t care about how they’ve arrived at a price or how it’s accounted for if I just want it be lower.
I say I want that price to be lower in order for me to buy and they say yes or no depending on if it’s worth it for them.
A more thorough discussion can bring down the offer they are willing to accept. You don’t have to care about the details to use them in a negotiation.
So why would you ever use the former approach? It’s strictly worse.
"Can you reduce bottom line by x" is like saying, "the documentation is a bit vague, can you improve it please?"
"Can you reduce line item n by x" is like saying, "there's no way the user could possibly intuit that the sampling period may be truncated to precisely hit the target value after the specified duration. Given that the programming language itself has no specification, this longstanding behavior must be explained clearly in the user docs."
Which response do you suppose will be more likely to lead to an improvement in the docs? It's probably the one that shows evidence you know your domain and how to assess quality there.
Of course, that analogy is the dream. In reality the car salesperson will come back with the reduced line item but with adjustments somewhere else. And it's hard to keep track of because, for example, how would an increase in the down payment affect your idea of the "total bottom line?"
This would be like a developer countering by saying, "Ok, but to clarify that section of the docs I added threading to the event loop." And then every single developer suddenly becomes some kind of darkworld saboteur, and you finally give in by saying, "I guess this is just how software gets made..."
Line itemisation is very silly for cars. All I need to know is how much does it cost me to get the car. I don't care that some of that money goes to person A and some of it to person B. Irrelevant to me!
But you would have already told them, "As long as I can afford..." So they'll spend the entire exchange trying to optimize for the maximum amount they think you can afford. They can try to convince you they've optimized when they really haven't, and they can try to pick a vehicle of a lower quality and persuade you that the price drop is the optimization.
If your response to every trick is to respond, "No, please try again to optimize for TCO," they'll eventually interpret your behavior as the equivalent of an ad-blocker. Then they'll either persuade you to reveal more data or move on to a customer who negotiates on their terms.
There is a difference simply because the results are different, whether due to the need to save face, behavior psychology, sales training, or more intangible factors.
What I'm saying is that either OP will accidentally give away information at the outset that virtually guarantees overpayment-- "Here's what I can afford"-- or robotically repeat a fair price calculated ahead of time and request TCO optimization until they are thrown out of the dealership.
It'd be like going to a shell game and requesting to pay $7 so your kid can have 10 guesses at where the shell is. Then when they ignore you, start asking what's wrong with that, I mean, those are the odds, no? I've even rounding up for that pure profit. Why wouldn't someone with a big crowd of good faith players want to accept that?
As to why you’d want to: if you ask someone to quote bottom line, and they do, then you can start picking their line items apart to go lower. Or if you quote high, you can use line items as justification for it to some buyers.