I agree there are only so many stores a mall can support. At that point the economically sound strategy is to stop building. It's not to layer on incidental complexity that provides lower returns at increased risks. It will only create problems syndicating investors for the project.
[1] Retail was still recovering from over-supply from the S&L pre-crisis and under-demand from the dot.com crash. At the low point of the cycle it could have been perfectly sensible to start planning a retail project on the prediction that retail could only come back and in the hope of timing the up-cycle correctly. But you would not have been building a mall. You would have been planning a power center.
Money people need things as simple as possible to package risk.
A family friend opened a bakery. Finding a storefront was an insane process, in some cases it’s apparent that the landlord doesn’t want to rent out the space, it’s been vacant for 4 years and rent is... aspirational.
Any real-estate pro forma will include a vacancy rate.
If a person owns ten storefronts on a block, discounting one lease sets a lower rate for the nine other leases.
If a business can't afford the lease, that is a problem with its cashflow and/or access to capital relative to its aspirations. Often the landlord can simply take depreciation and meet its IRR targets. Particularly if the landlord is unleveraged. The building and land aren't going anywhere. That's the nature of real property as an asset class.
Finally, there is the issue of investment. There are big pots of money for each use...investors who invest in retail, office, and residential. The pots of money for non-conventional projects are much much smaller. One reason is that with mixed use, there are three economic cycles that have to be timed. The retail, office, and housing sectors don't cycle in lock-step.
Your good ideas in real-estate only have legs if you have the money to make them happen. And the time. You're talking a several hundred million dollars and a decade optimistically down the happy path. More likely more since most mall land is already malls and you will have to acquire multiple parcels from multiple owners without them being the wiser and holding out for premium prices.
[1]: Retail uses produce so much income that it makes sense for them to sit vacant for many years instead of converting to some other use.