When I was finishing my planning degree roughly a decade ago, I consumed hours of Dan Bell’s Dead Mall Series on YouTube while hunkered down in the GIS lab late into the night. This on-again, off-again preoccupation with dead malls was rekindled a few weeks ago when I visited Sunrise Mall in Citrus Heights, a suburb of Sacramento.1
It’s the prototypical dead mall: Built in 1972, it seems like it was last spruced up in either the late 1990s or early 2000s. The vast parking lots sit empty, strewn with litter and dead leaves. Inside, underwatered tropical plants hang on by a thread. Of the five original anchor tenants, only JCPenney survives. The rest of the mall is filled with low-margin businesses: a local clothing store, a cellphone repair shop, a Bitcoin ATM.
The macabre charm of the dead malls is undeniable: As recently captured in Backrooms, there’s something entrancing about neglected, surplus commercial space. Perhaps we take some strange sort of moral comfort in a project so grand failing so spectacularly. Or maybe it’s the faint aftertaste of nostalgia to be found in aesthetics that are 30 years out of fashion.
But wait: Why are all of these dead malls still around? And why in the world can you walk around inside of them?
It’s more complex than you might think.2 Yes, yes, the proximate cause of any given dead mall was some mixture of the construction of newer malls, the decline of traditional anchor tenants, and the rise of e-commerce. But retail businesses go under all the time. For the most part, they don’t blight the landscape for decades.
In 2017, Toys “R” Us filed for bankruptcy: it was a retail mass extinction event in which 725 stores closed and 30 million square feet of space suddenly came online. A mere six years later, 90 percent of this big-box retail space is back in use as something else. Look up your hometown Toys “R” Us location. What is it today? Mine is a Best Buy.
The sheer adaptability of allegedly “fragile” suburbia has even become a meme. Over on Reddit, entire communities exist to document how conspicuous former Pizza Hut, Blockbuster, and Taco Bell locations have turned into urgent care centers, carpet stores, and endless local restaurants.
Yet malls infamously sit empty, sometimes for decades. Growing up, my local mall was Lexington Mall.3 By the mid-to-late 1990s, most of the anchors had left. Even so, the almost entirely empty mall remained open until 2005, when the last anchor tenant—Dillard’s—departed. It then sat empty and shuttered for another six years, before being partially demolished in 2011.4
Stranger still, you can often still go inside dying malls, even in an advanced state of deterioration—a sort of baby’s first urban exploration experience.
On a recent road trip, I stopped by the Barstow Mall. It was rough. The anchors had departed over 20 years ago. One anchor space had been sectioned off and turned into government offices. The mall was otherwise empty, except for a single tenant.5 Nonetheless, the entire interior corridor space remained open and neglected. At the far end of the mall, there was even a homeless encampment.
What in the world is going on here?6 How do malls remain open long after the writing is on the wall?7 Wouldn’t it make sense to move heaven and earth to buy out the remaining tenants and shutter these hulking liabilities? Why are dead malls a thing?
The actual answer is that dead malls are the physical embodiment of a spectacular breakdown in private governance.
In the mad dash of the mall-building boom of the 1970s and 1980s, mall operators were desperate to attract the anchor tenants needed to draw in customers. To attract chains like Nordstrom and Sears, they gave anchor tenants veto authority over nearly every aspect of mall governance: hours, tenants, signage, parking ratios, site design, and even whether the property had to stay an enclosed mall at all.
It must have seemed smart at the time, signing away “free” covenants, reciprocal easement agreements, exclusive-use clauses, operating covenants, and consent rights controlling every single aspect of how the mall is governed to four or five different companies, in exchange for a bustling mall.
In the good times, none of this really mattered. Yes, it must have been inconvenient to get the sign off from the various companies on every small programmatic change. Maybe Bloomingdale’s would balk at developing part of the parking lot, or Boscov’s would veto adjusting the mall’s hours. No bother: As long as the anchor tenants were happy, such spaces could easily be filled with a tenant that meets their standards.
Then came the bad times. When the anchors started going under, this thicket of rights and obligations made it nearly impossible for malls to quickly adapt.8 And in most cases, it was the anchors that survived the longest that blocked change.
Consider the case of Lord & Taylor v. White Flint. In 1975, the operators of White Flint Mall in the Maryland suburbs of DC signed a lease with Lord & Taylor agreeing that they would operate an enclosed mall until 2042 (yes, really) and give the company veto rights over any major changes to operations. By 2013, Lord & Taylor was the last tenant in an otherwise dead mall. The plan was to replace most of the mall with a mixed-use development, while preserving the Lord & Taylor. A win-win, right?
Except, by this point, Lord & Taylor was barreling toward bankruptcy itself, and the private equity firm that had purchased it in 2006 smelled blood in the water. It did not go well for the mall operators, who were ultimately forced to pay out $31 million in restitution. Today, the site of the old White Flint Mall sits empty—except, ironically, for a shuttered former Lord & Taylor location.
This story is only exceptional to the extent that it got so far along. In case after case, the scrap pickers that own companies like JCPenney and Macy’s commonly block or delay efforts to make even the smallest adjustments to dead malls—that is, unless the mall operator is willing to make a hefty payout. In many more undocumented cases, redevelopment proposals likely never even get off the ground.
All of this is a big problem for YIMBYs, who have put a lot of stock in mall sites as an opportunity to build a lot of new housing.9 Even if the zoning is fixed and the permitting processes are streamlined, viable redevelopment sites may sit empty as long as these private rules remain in place. If it takes “remarkable” and “unheard of” deals to turn dead malls in ultra-expensive Cupertino into mixed-use communities, don’t expect this approach to scale.
Remarkably, mall operators and commercial landlords still have not wised up. To this day, such restrictions remain standard features of mall leases for highly coveted tenants, and they frequently derail mixed-use redevelopments. You are almost certainly underestimating the scale of this problem, as these disputes nearly always unfold in private.
This issue is hardly unique to commercial sites. Indeed, these cases are among the most easily solved, to the extent that the solution is just a payout to the holdout. In neighborhoods across the country, covenants, conditions, and restrictions (CC&Rs)—the contractual basis for homeowners associations—now serve broadly the same function, strictly defining the narrow details of what can and can’t be built on any given residential lot.10
These private regulations are only voluntary in a very theoretical sense: In many cases, cities condition permits for new housing subdivisions on the adoption of CC&Rs, such that roughly two-thirds of new homes are subject to them. Harried homebuyers rarely have the time to meaningfully review them before purchasing a home, and they never have the right to modify or opt out of them.
Worse yet, these modern covenants usually never end: Prior to the 1910s, courts generally enforced the common law norm of the “rule against perpetuities,” tossing out covenants that were adopted without any end date. As a result, most covenants lasted 30 to 40 years, and subsequently needed to be proactively renewed to continue. In the 1910s—partly in deference to the rise of racial covenants—courts gradually stopped enforcing this rule.
As a result, the modern American city is blanketed in what Yale law professor Robert Ellickson has dubbed “stale covenants,” or restrictions on property that no longer serve any conceivable purpose, yet they remain in full effect, constrain the productive use of property, and are prohibitively difficult to remove.11
What can be done? While I’m dispositionally laissez faire—I did write a whole book on abolishing zoning—this is the sort of market failure that you need regulatory guardrails to fix. In the spirit of not carefully documenting a problem and then abdicating any responsibility for proposing solutions, here are three ideas.
First, in the clearest cases, certain types of covenants should just be suspended. Do we need to wholesale suspend the CC&Rs for brand-new HOAs, or take away Auntie Anne’s right to not have any spaces near them in the mall leased to Wetzel’s Pretzels? No. But legislatures have historically had no qualms about rendering covenants banning American flags, solar panels, or racial groups unenforceable.
Here in California, a key part of the accessory dwelling unit (ADU) renaissance has been state laws suspending CC&Rs that banned them.12 This year, we are hoping to do something similar for townhouses and small-lot subdivisions.13 Last year, the state also provided a pathway to remove restrictive covenants banning residential developments on commercial properties. More states should follow.14
Second, we should put additional guardrails around the various legal instruments that frequently ensnare reasonable redevelopments.15 For the most part, covenants and similar agreements should be time limited to 30 or 40 years, with automatic expiration barring affirmative renewal.16 Crucially, they should be transparent—that means putting them in machine-searchable public databases. All too often, planners and developers don’t even know these constraints exist.17
If this seems radical, it isn’t: Scotland abolished feudal land tenure in 2000 and, three years later, put a sunset on the private restrictions the feudal system had spawned. As part of this, they established a mechanism for clearing out “real burdens” older than 100 years. Unopposed applications are approved as of right without much fuss. As California itself slips into a sort of neofeudalist state of untaxed, inherited estates, perhaps it could learn from this approach.
Finally—and here, I will admit that things get a little more extreme—courts should lean into the doctrine of changed conditions more. Without requiring legislative change, courts could recognize that when reality changes, so too must out-of-date agreements. Courts were historically more comfortable tossing out incoherent restrictive covenants when new transit lines opened. Does a contract on mall hours written in 1976 really deserve unlimited deference in 2026?
There is nothing stopping any of this today. Of course, his mechanism probably won’t be much use for recent agreements, active leases, or dispersed land ownership. But in cases where such flexibility could, on the margins, provide a rationale for bringing contractual fictions back in line with reality, courts should embrace it.
Of course, there is always risk of overcorrection: the reality is that, always and everywhere, social order depends to some degree on private governance. At their best, malls and HOAs provide the supplemental services and rules and governments can’t, or shouldn’t, provide.18 The most recent iteration of the mall—the outdoor lifestyle center—seems to be a step toward a better balance, with consolidated ownership and limited leases.
In the meantime, go enjoy one last slice of Sbarro or bowl of Manchu Wok. Get your ears pierced at Claire’s and buy a cringe graphic tee from Hot Topic. Power walk through the corridors, and do donuts in the parking lot. With any luck, the dead mall will soon be an oddly comforting memory.
I was out there in pursuit of one of the last Barnes & Noble locations near me with a full-sized multimedia section—for the Criterion Collection summer sale, naturally!
And probably slightly overstated. As convincingly argued in Meet Me by the Fountain, Alexandra Lange’s excellent history of malls, malls remain far healthier than the popular imagination might lead you to believe.
True to the “malls aren’t really dead” contrarian thesis, the main killer was the rise of Fayette Mall across town, which remains very much not dead.
In 2013, it reopened as a megachurch. Local property tax bean counters could not have been happy about this development.
A great little retro video game store, and my reason for visiting the mall.
If there is any theme to my past few posts, it is that urbanists should spend less time judging cities/neighborhoods/buildings and more time trying to understand how they work.
Or in the case of the Barstow Mall, the graffiti on the wall.
Signing off on one of the other anchor spaces turning into a call center probably wasn’t a top priority for Goody’s staff as the company started bleeding to death.
For reasons detailed below, this is more so now an issue for YIMBYs outside of California, as AB 1050 (2025) suspended restrictive covenants that outright ban residential uses on commercial lots.
As a matter of course, these CC&Rs usually prohibit anything other than a detached single-family home, and they usually mandate that lots and homes meet a minimum size threshold.
Shout out to now-Congressmember Laura Friedman.
Shout out to Senator Anna Caballero.
Shout out to Assemblymember Nick Schultz.
Note that changes to federal underwriting standards could easily solve a lot of this problem going forward.
I’m looking at you, perpetual conservation easements.
This is yet another problem with RHNA, California’s housing fair share mechanism. Often, cities allocate their permitted housing to sites subject to legal constraints that effectively ban redevelopment.
As I argue in my book, the social function of deed restrictions is to satiate the demand that many households have for regulations that go beyond true health and safety considerations; when such demands are not met through private governance, they are met through public regulations that do far more harm.
There's a similar issue with intellectual property, where both the sheer length of copyright terms and the uncertainty over what's in or out of copyright restrict the reuse of our cultural inheritance. Consistently and thoroughly applying the general principles that legal/contractual restrictions of these types:
-- sunset after a reasonable period of at most a few decades and certainly much less than a typical human lifetime (not "life of the author plus 70")
-- must be centrally, accessibly registered to be enforced
-- will not be enforced if they serve antisocial goals (like racism, solar panel restrictions etc)
would probably do a lot to de-sludge economic and cultural activity generally.
We have a similar case in Los Altos, with a 1950s parking easement agreement among seven commercial properties. Several of the properties have vacant office buildings that cannot find tenants at any rent. One or two of the properties are condo-ized offices, with 50 or 100 owners. There is virtually no way to find enough of the owners of the condo offices to revoke the easement, so acres of land, that should be over $10 million an acre, are sitting empty with vacant parking lots and empty offices. I've spoken with two developers who banged their heads against the wall fruitlessly, trying to figure out a way to develop one or another of the properties as housing.