But to answer your specific question, the most obvious negative externality is increased rents for other people, because an apartment that could be being rented out (increasing supply) isn't being rented out.
Having lots of vacant residences also potentially increases crime, is bad for neighborhood maintenance (we have a huge problem here in NYC of sidewalks not being shoveled in front of vacant properties when it snows), it gives pests a potential home, and increases the potential to damage surrounding residences. Vacant residences are more likely to fall under disrepair and then catch on fire or start leaks, which can damage nearby or underneath residences. If you're touching one then it's bad for your heating/AC bills, because your heat/cold is sapping through the walls into the un-conditioned apartment. Even just simple things like having fewer eyes to watch out for crime and call the police if a burglar or robber is spotted.
There's a huge host of problems associated with having long-term vacant residences in your neighborhood.
What you are describing is not an “externality” in the way that economists use that term.
Owning central land and not using it means that millions of people have to go further to get to their destinations. It also means that construction effort and money is going into a black hole and not turning into supply. Enough of this together makes a powerful negative externality.
Because of high rises and land reclamation, this is a lot less true than you'd think.
Can you create new land?
No. They are completely different things.
https://squattinglondon.files.wordpress.com/2018/01/ominous-...
and it seems obvious to me why that's bad and public policy should not promote it.
As a public policy, along the lines of taxing vacant properties, we could also simply permit any vacant or unused property to be lived in and used by anyone who came along.