OP is assuming the oversubscription is due to counterparties being on the hook for gold they are obligated to deliver at a certain time (this is commonplace with commodity derivatives, you simultaneously create contract to buy a certain amount of a commodity at a certain time in the future + a contract to sell a certain amount at the same time in the future - they cancel each other out), not that the entity holding the gold fradulently issued more claims than they could deliver on.
So OP is saying that if more gold was redeemed than expected, the counterparties to those claims would be forced to buy gold at whatever the price it ultimately settles on to fulfill the contract.
Of course the problem is that there is such a thing as counterparty risk. If delivering the contractually obligated gold bankrupts an individual counterparty, or is physically impossible in aggregate (eg more claims than gold exists) then the contract won't be fulfilled at least for someone.
But there is a lot of complexity here because not all futures are settled physically (they can be cash-settled) and there could be mechanisms in place to manage counterparty risk at various levels, like at the exchange, eg https://www.cmegroup.com/education/articles-and-reports/coun...
I wasn't able to read the article because of the paywall but stuff like this is why regular joes tend to think there are bigger problems with gold or finance in general than there actually is. "There is more paper gold than physical gold" - regular Joe thinks there is a problem. "Actually the paper claims' settlement are subject to various conditions, typically closed out before taking delivery, and there is a complex system in place which does its best ensure that anybody who actually wants to take delivery or settle their contract gets it settled as expected" - regular Joe lost interest and went back to doomscrolling.
The problem here is obviously liquidity management. You could think of the vault as a battery that charged and discharged.
If you drain and charge the battery at the same time, there is no net change in the charge level. Theoretically, you wouldn't need the battery to begin with.
Now add time lags. Now discharge may happen before charging. The net change is zero over the full duration of the cycle, but the battery is being drained temporarily and it is changing its state all the time.
If the battery is not large enough, the system will shut down. This in itself is not necessarily a catastrophe, but you will have to pay damages over the shutdown duration for the consequences of that shutdown. This will probably ruin profitability of the counterparty.
So OP is saying that if more gold was redeemed than expected, the counterparties to those claims would be forced to buy gold at whatever the price it ultimately settles on to fulfill the contract.
Of course the problem is that there is such a thing as counterparty risk. If delivering the contractually obligated gold bankrupts an individual counterparty, or is physically impossible in aggregate (eg more claims than gold exists) then the contract won't be fulfilled at least for someone.
But there is a lot of complexity here because not all futures are settled physically (they can be cash-settled) and there could be mechanisms in place to manage counterparty risk at various levels, like at the exchange, eg https://www.cmegroup.com/education/articles-and-reports/coun...
I wasn't able to read the article because of the paywall but stuff like this is why regular joes tend to think there are bigger problems with gold or finance in general than there actually is. "There is more paper gold than physical gold" - regular Joe thinks there is a problem. "Actually the paper claims' settlement are subject to various conditions, typically closed out before taking delivery, and there is a complex system in place which does its best ensure that anybody who actually wants to take delivery or settle their contract gets it settled as expected" - regular Joe lost interest and went back to doomscrolling.