Honest question: if we don’t know the shape of the curves, if we can’t observe them, and if they shift around according to unknown dynamics, what good are they?
The supply-demand curve logic is an excellent argument to point out to people that, based on some basic and obvious axioms, their intuitions are dead wrong. They are justified with just that observation.
Take, eg, a crisis. Price caps in a crisis are a terrible idea that are about to make the situation much worse. There was a shortage, now there will be a bigger shortage. That seems to be unintunitive to most people but anyone who has had some basic supply-demand curve training can be talked out of imposing caps.
Economists have been idly trying to figure out a way to pitch the idea that wages are a price and therefore subject to supply and demand model for decades. No luck so far; it is difficult to get a man to understand something when that thing is his salary.
Also, it is more interesting to try and find places where supply-demand doesn't work and then figure out why not. Usually the answer is regulation that creates shortages or overproduction.
If you want a specific example; masks in COVID. There were price caps that guarantee we'll be caught short of masks the next time a respiratory crisis rolls around. It isn't like Bill Gates didn't see it coming, if there was money to be made he'd have a strategic mask reserve ready to go for the next one. But no. Price caps. There was even an article about a US mask producer who was refusing to expand production because he knew how it'd play out, unless my memory deceives me [0].
[0] EDIT Found the article - https://www.wired.com/story/surreal-frenzy-inside-us-biggest... . Bowen doesn't mention price caps, but you can see he is no stranger at all to how that sort of medical crisis plays out. He'd be the sort of person in a great place to make decisions about what sort of surge capacity to have on hand and how to plan for it and he is mainly responding to basic price signals. I still suspect that he'd be sensitive to the eminent domain risks under crisis conditions.
If there is a crisis for a critical good, the only solution that works is to throw the market out altogether and create a temporary planned system of rationing. If five people are on a desert island with five watermelons, allowing the billionaire to eat all five because she has more money is not going to help. Similarly, if a nation can't produce or import enough food (e.g. in war time), you not only cap prices on food, you ration all the available food to try to make sure no one starves to death even if everyone is left hungry (or in even worse crises, allow people who wouldn't survive anyway to starve so that those that have a chance can get some rations).
In reality, even for much milder crises, finding extra supply if prices are allowed to rise indefinitely happens rarely. Letting prices rise to any value is typically just a way to make sure the existing suppliers become filthy rich, it rarely does anything else.
Of course, there are limits. If the government capped smartphone prices at 1$, we wouldn't actually get 1$ smartphones, we'd see the supply plummet to 0.
But the reality is the inputs need to be known to make the curves. Supply curve is 'easy' as you know what it takes to build X number of items. So you can make that curve easily. The hard part is the other curve. How do you know I can sell X of an item. Is there demand there? 'utility' 'happiness' and other things like that make up that curve. But how do you measure those inputs? If you have been around awhile you probably have a good idea where your curve is by seeing it in action, but the inputs for it you are usually guessing. But starting out? Not really. What really bakes someone's noodle is the supply curve can affect the demand curve.
You want to get as close to MR=MC as you can. But that is not always achievable. As you point out the curves move around. But that is on you as a business owner to make sure your prices are in line with your demand.
Will estimates always be right? Of course not. But nobody has perfect knowledge about anything. That doesn't make analysis based on estimates useless.
Even with order books though, they just provide point-in-time snapshots. It’s well-known that buyers will get a better fill than what’s shown in the book if they break their order up over time: that’s why it’s normally to trade VWAP.
These are definitely the exception rather than the rule. I work for a financial services firm and I can confidently say we have no idea what would happen to demand if we cut our fees.
I'd say modelling trading impact and trying to optimise around it is actually quite intense modelling of supply and demand at times.
But have you actually tried to establish elasticity and run experiments to that end?
In all my time in industry, I have mostly not seen efforts to do that, but when people did that it actually was quite successful in increasing revenue.
Fixed the link - thank you.