A shortage of a good means that there if there can be pending buy and sell orders in a market book, then there are a lot of pending buy orders, but just not for the price at which sellers are willing to sell.
What SV is saying that there are a lot of people who would buy a lot of goods, if they can be produced cheaply (and it isn't because they're poor, it's just because the consumers don't value that good that much). If Netflix raised their fees to $90 per month then it means that a lot of people would cancel their subscription (and not all would do because at $90 they will be unable to meet their rent).
If Netflix magically halved their wage bill they would likely keep their price the same since how much a Netflix devops engineer gets paid doesn't change how much a family in Peoria is prepared to pay for a Netflix subscription.
Just because a service exists at some (arbitrarily high) price point doesn't mean society isn't suffering from a shortage of that service.
If Family Doctors could charge $10K/hour, that most certainly does point to a shortage.
His point was, offer Family Doctors $5/hour and see how many family doctors show up, which is 'equivalent' of what is happening now.
I'm pretty sure the word shortage already implies that there is a shortage at the market price, not at an arbitrary price.
Of course not every market is a large efficient market. If you're talking about a price-regulated market, by all means, discuss shortages. Another good situation where discussing a shortage would be appropriate is markets that have very or perfectly inelastic supply over a relevant time period. Say, a vaccine that costs $1 but that takes 6 months to produce due to the biology involved, combined with a major epidemic in the relevant disease. It's fair to call that a shortage.
Constraining supply increases salaries precisely because it takes supply so long to incorporate information.
On top of that, supply and demand is a pretty leaky abstraction, I don't expect it to be very accurate in such a nonideal market.
For any given skill there is a labor pool with some participation rate, of people who currently have that skill, such that it is accurate to say that the market has a shortage.
Answers to the shortage must solve for either the size of the labor pool, the rate of participation, or the amount of people who have that skill.
Those answers vacillate between importing labor to solve the problem now, or forcing the market to educate people by blocking imported labor.
The problem with importing labor is that it reduces market incentives to educate people. The problem with not importing people is that it means you'll have a labor shortage until the market adjusts and people re-educate.
Easy answers, on either side, are snake oil.
"Shortage at the market price" seems to be a somewhat incoherent concept. The market price is the price point at which the supply and demand curves intersect. Other than a discontinuity in either supply or demand such that the curves pass without intersecting (which makes the market price ill-defined), demand equals supply at the market price.
We're pretty comfortable talking about shortages in consumer markets when we would buy things at the stated price, but can't because it is sold out (eg Pixel phone shortages), so I'm not sure why we can't say a similar thing about labour.
That can just be a current price below the equilibrium market price; I'd argue that a shortage is a specialized instance of that resulting from a transitory decrease from long-term supply (or increase from long-term demand) that is expected to revert to normal.
But if you need annual supplements of labor for the same field over a period of decades, that's not a labor shortage, its price being artificially suppressed (and inhibiting market signals that would develop domestic supply faster.)
Theoretically, couldn't it also be that demand is consistently growing faster than anticipated (even after whatever corrections are made upon discovery that it previously grew faster than anticipated)?
Google literally does offer $1 million salaries for the right positions, the problem is that most of the people who could make that work have already been recruited.
Which is why they pay interns six figures to show up for what amounts to a long job interview, to make sure they get all the new talent the moment it's minted.
If you make it more difficult to recruit people from untapped populations abroad, they won't pay lower tier people more, they'll open offices aboard to capture their acceptable talent.
So yes, there is a shortage of talent that meets the bar. While everyone trained as a book keeper can almost certainly fill any book keeper role, not everyone trained as an engineer can actually fill any role. Not just because some people are shitty engineers, it's because engineering involves spending long periods of time learning domain knowledge that isn't transferrable, even setting aside the fact there are literally thousands of types of engineer.
FWIW, my rant was not intended as a rant against immigrants, just a rant against people that spin their personal or corporate financial interests as a national crisis.
The company is heavily automated and very profitable, which means they can attribute those profits to the humans largely responsible for them with accuracy that isn't often seen outside finance. The bonus structure is extremely kind to the highly productive.
It is a good point that the market can't achieve the impossible. C++ bootcamps probably wouldn't spring up in your hypothetical ancient Rome and offering millions of dollars today for employees with skillsets that modern humans can't achieve wouldn't work either.
The question is whether you or not the current bar of a "qualified candidate" is impossible. I would say no. Given the right financial incentives, the candidates will exist.