Because the supply of land is essentially fixed, land rents depend on what tenants are prepared to pay,
That doesn't make any sense. If more people want to live in a particular place, rents will increase.
And if rents increase, then obviously tenants are prepared to pay more. If they aren't willing to pay increased rents, then they don't really want to live there more than another place. I.e. your statement is entirely consistent with the statement that you are criticizing.
rather than on landlord expenses, preventing landlords from passing LVT to tenants.
Only if there's somewhere else that tenants want to be, that also has lower rents. The real estate boom in San Francisco shows that when people with lots of money really want to live in a particular place, rents rise.
I think you missed the point here. Presumably landlords are already charging what the market will bear. How does a Land Value Tax (LVT) change what the market will bear? As it's a tax on landlords, obviously it has no direct effect on tenants. The tenants are not prepared to pay more rent as a result of the LVT.
If rents will go up by \$100 a month without an LVT and \$100 a month with an LVT, then landlords can't pass the LVT to tenants. That's true even if the \$100 is enough to pay the LVT. Without the LVT, the landlord would have raised the rent and kept all of the money (minus income taxes, etc.). With the LVT, the landlord pays the LVT and may still have to pay more income taxes (depending on the deductibility of the LVT).
The only way that an LVT can influence rents from the landlord side is if it causes the number of units available for rent to either increase or decrease. In the long term, that happens. Landowners can choose to build more or fewer units on their land. But in the immediate term, the number of units is fixed. A landowner can't take a single family home and instantaneously convert it into a hundred-unit apartment building, even if there is sufficient land to support that. Thus the LVT does not change unit availability or the rent that tenants are willing to pay.
Now, it is possible that the LVT enables spending that tenants find valuable. And that spending causes demand and therefore rents to increase. But that spending would be just as valuable to the tenants if it were financed by a different tax with less impact on landowners. Of course, tenants might be negatively impacted by that tax, e.g. if they pay it. But those effects are aside from the impact of an LVT on landlords.
The LVT itself does not increase rents in the short term. It may increase rents in the long term if it drives away landowners by making some places economically infeasible to rent. But that assumes that landlord profits are less than the LVT. Or landowners may choose not to expand the number of units in the long term. Or the LVT may finance a benefit that increases the value of the units to tenants.
So the landlord can only pass on the LVT if tenants benefit from the spending financed by the LVT, shifting the demand curve such that tenants will pay more. So the more that the LVT finances benefits for tenants, the better able the landlord will be to pass on the LVT.
If those things don't happen, the landlord can't pass on the LVT. The landlord can only get the regular market rent. That may be an increasing market rent, but the increase is not due to the LVT. It's only determined by movement in the tenants' demand curve, which is not directly impacted by the LVT.