I'm not sure about the relationship between "economics simulator" and "D&D doesn't make allowances for changes in fashions of gem cutting".
Because fashion is one of the factor of price for gem (alongside with rarity) ? Assuming a static demand everywhere, everytime, quickly breaks verisimilitude as in the examples I provided. I mentionned fashion to illustrate the market value diffrences in a single place over time, but I could also have mentioned that your 50gp ruby as determined by the tables when you find a treasure ("objective" value) is different from the market price in the plane of Earth (just pluck it from ground of the Ruby Plain at no cost, they aren't fashionable as status symbol since any peasant xorn can have one to eat, and them being a status symbol is the main reason to have gem being pricey) and in Waterdeep (regular price). Much like spices being the same price in the rulebook doesn't mean it's the same price in Waterdeep (many rich merchants, spices fashionable, high price), in a hamlet (demand focussing on bread and meat, not spices, sure it makes food taste better but seasoning is totally unnecessary, average price) and in a place at the other side of the world where the spice grows on trees (nearly free). The rules prices don't reflect market prices. They might reflect market price in the home town of the PC the day they start adventuring, but after that, either they are static (breaks quickly) or dynamic, even if slowly moving, and it's the job of the GM to account for it and provide the player with a market value appropriate to the time, place and circumstance where they try to buy or sell things, thus determining the price for their 2cp loaf of bread, 15gp sword and 50gp ruby.
To put it another way:
Case A: If prices are static and disconnected from economy, I can buy a ruby listed in the rules as a 50gp ruby, use it to create 50gp of ruby dust for spellcasting purpose, at the market price of 50gp. Pro: it's damn easy! Con: it relies on static market price, which are quickly causing disbelief around the table as the concept of static price confronts real life reasoning.
Case B: If price are dynamic and the spell requires market value spending, as a PC, when I need to cast the spell, I need ruby dust, in an unknown and variable amount, but I just have to be 50gp poorer at the end of the transaction for magic to happen.Pro: it's simple. Con: it relies on atypical assumptions on the working of magic.
Case C: If price are dynamic and the spell requires a "50 gp ruby", then... as a PC, I can't know if a ruby I buy for 50gp is a 50gp ruby, it could be a 10gp ruby that has a specific history or kardashian markup, or I could be in the wedding seasons and ruby are highly sought-after, or it could be a 1,000gp ruby and in my last adventure I went back from the Earth Plane with enough rubies to depress price a lot. So I am having a hard time buying a 50gp ruby (I can't rely on weight since the spell doesn't mention how much dust I need). Also, since a "50gp-ruby-in-the-rule" depends on size and cut, sometime a pinch of ruby dust (coming from a very sought after cut) will make the spell work, and sometime I will need much more (since it was a rough ruby that was ground, or worse, my dust is a by-product of creating a proper 50gp ruby with no rubiosity left in the dust). And seeing the dust, I really can't tell from what ruby it was made so I really can't know before casting the spell if it will fail or not.
I really prefer case B, despite the drawback, because the other situations sound horrible (in real life for A, in-game for C).