Parking real estate repriced the moment commuting collapsed, years before office rent rolls admitted the same shock. The reason is structural. Parking income is largely daily and mark-to-market: a transient parker or a monthly permit-holder can walk away this month. Office rent is contractual, locked into five-to-ten-year leases that only reset as they expire. When hybrid work cut the daily commute, parking revenue fell in real time while office rent kept accruing on paper. Parking was not a laggard to the office story. It was the leading indicator, and operators who read it early saw the demand shift before it reached the rent roll.
Key Takeaways
Parking income is daily and mark-to-market, so it repriced remote work immediately, while office rent lagged behind on multi-year lease terms.
U.S. office occupancy sat near 51.9% across 10 major metros in April 2024, per NAIOP citing Kastle Systems, a level parking demand had already absorbed years earlier.
Average downtown office vacancy rose from 10% to 16.2% between 2019 and 2022, per NAIOP, but parking felt the commuter loss first because its revenue turns over daily.
A garage carries high fixed operating cost, so a daily-demand revenue drop hits net operating income with severe operating leverage, as the worked example below shows.
Structured parking costs roughly $20,000 to $40,000 per space to build, per NAIOP, and a national median near $29,900 per space in 2024, per WGI, which makes overbuilt parking an expensive misread of demand.
Why Did Parking Real Estate Reprice Before Office Did?
Parking real estate repriced first because its income is mark-to-market and its demand is daily. A monthly permit or a transient ticket can vanish the moment a commuter stays home, so parking revenue tracks actual attendance in near real time. Office rent, by contrast, is contractual and lagging, held in place by leases that only reset on expiration.
This is a timing difference, not a severity difference. When the daily commute collapsed, parking felt it in the same quarter. Industry reporting captured the scale: office REIT Equity Commonwealth booked $1.5 million in parking revenue in the third quarter of 2019 and $736,000 in the same quarter of 2020, a 53.2% drop, per Bisnow. Office rent from the same era did not move that fast, because a tenant on a 2026 lease expiration kept paying 2019 rent regardless of whether anyone showed up. The office repricing arrived later, as leases rolled and vacancy climbed. NAIOP reported average downtown office vacancy rising from 10% in 2019 to 16.2% by 2022, a slow adjustment compared with parking's overnight one.
The operator lesson is that mark-to-market income is an early-warning system. Parking told the truth about demand while the lease-bound office rent roll was still reporting a world that no longer existed.
How Does Parking Generate Income, and Which Revenue Type Broke First?
Parking generates income through three main streams: contract (monthly permits), transient (hourly and daily), and event or special-use revenue. Contract parking is the commuter subscription, transient is drive-up demand, and event revenue captures stadiums, venues, and peak-hour surges. Remote work hit the commuter streams hardest, because the daily office trip was the thing that disappeared.
Revenue type | Source | Demand driver | Remote-work exposure |
|---|---|---|---|
Contract (monthly) | Recurring commuter permits | Days per week in the office | Highest: hybrid workers stopped buying full-time permits |
Transient (hourly/daily) | Drive-up parkers | Foot traffic, retail, visitors | High in office districts, mixed where visitors returned |
Event and special use | Venues, stadiums, peak surges | Events and destination demand | Lowest: unrelated to the daily commute |
The contract stream broke first and hardest. A hybrid worker in the office three days a week has no reason to hold a full-time monthly permit priced for five, so operators watched recurring subscriptions convert to occasional daily use or evaporate. NAIOP documented the operator response: some raised rates to offset lower volumes, and parking revenues recovered in places even as parking volumes stayed down. Transient revenue held up better where downtown visitors returned faster than workers. Robert Dunphy, writing for NAIOP, noted that some markets had recovered more than three-quarters of nonresident workers by mid-2023 while the weakest Western downtowns hovered just above 50%. Event revenue, tied to destinations rather than commutes, was the most insulated of the three. This is parking income, the same other income line that underwriters often treat as an afterthought, behaving as a precise sensor for how a location is actually used.
What Does a Garage's NOI Look Like Before and After Remote Work?
A parking garage carries high fixed operating cost, so a decline in daily-demand revenue drops through to net operating income with heavy operating leverage. Structure debt, staffing, insurance, and maintenance do not shrink when parkers stay home. The worked example below holds expenses roughly fixed to isolate what the demand shift alone does to NOI.
Take a 600-space downtown garage. The inputs below are illustrative, chosen to show the mechanics, not drawn from a specific property.
Before remote work:
Contract: 400 monthly parkers at $250 per month times 12 = $1,200,000 per year
Transient: 200 spaces at $18 average daily rate times 300 revenue-days = $1,080,000 per year
Gross revenue: $2,280,000
Operating expenses (largely fixed): $760,000
NOI: $1,520,000
After remote work, with contract demand halved and transient down by 40%, even after a rate increase to $20 daily:
Contract: 200 monthly parkers at $250 times 12 = $600,000 per year
Transient: 120 spaces at $20 times 300 = $720,000 per year
Gross revenue: $1,320,000
Operating expenses (still largely fixed): $740,000
NOI: $580,000
Gross revenue fell 42%, but NOI fell 62%, from $1,520,000 to $580,000. That gap is operating leverage. Because the cost base barely moved, the revenue decline landed almost entirely on the bottom line. This is why parking values moved so fast: the income that repriced was also the income the whole valuation sat on. Underwriters who model parking as a stable, fixed contributor miss this. The mechanism is the same one that makes NOI fragile whenever an income assumption is treated as durable, a failure mode covered in where underwriting models go wrong on NOI.
What Does Early Repricing Mean for Property Types and Land Value?
Early repricing means parking exposed a demand assumption that had been baked into decades of zoning and site design. If commuter parking demand can fall by half and stay there, then minimum parking requirements built for a full-attendance world overproduced a low-value use on high-value land. The economist Donald Shoup spent a career making this case: minimum parking requirements are set from studies of demand for free parking, not market demand, so they systematically overbuild.
Shoup, the UCLA urban planning economist who died in 2025, argued that parking mandates can raise development costs by more than ten times the impact fees for all other public purposes combined. Remote work turned that critique from theory into a balance-sheet event. Cities responded: as of 2024, more than 3,700 municipalities across 22 countries had reduced or removed parking requirements, and over 100 had eliminated them entirely, per the Parking Reform Network. A 2025 U.S. Department of Transportation analysis estimated that removing parking minimums in Colorado would enable 71% more homes in transit-oriented areas and 41% more homes overall in the urban areas studied.
The reframing is direct: a garage or surface lot competes with the buildings that could occupy the same footprint. When parking demand falls, the floor area ratio that zoning permits on that parcel becomes worth more as housing, retail, or mixed use than as stalls. NAIOP noted that existing downtown parking, costly to replace at $20,000 to $40,000 per space, is now a resource to convert rather than expand. Shared parking, which serves office, retail, and restaurant peaks with fewer combined spaces, does the same work. Parking priced in remote work early, and in doing so it repriced the land under every overbuilt lot. The same property-types logic, that pricing an asset on last cycle's demand assumption is the surest way to misread it, drives the shift in why data centers underwrite nothing like offices.
Frequently Asked Questions
Why did parking income fall faster than office rent during remote work? Parking income is daily and mark-to-market, so it drops the moment commuters stop showing up. Office rent is contractual and set by five-to-ten-year leases, so it only resets on expiration. Parking repriced remote work in the same quarter it happened, while office rent adjusted years later as leases rolled.
What are the three main types of parking revenue? Parking revenue comes from contract permits (recurring monthly commuter subscriptions), transient parking (hourly and daily drive-up demand), and event or special-use revenue (venues, stadiums, and peak surges). Remote work hit contract and transient commuter streams hardest, while event revenue, tied to destinations rather than daily commuting, held up best.
Why does a small drop in parking demand cause a large drop in NOI? A parking garage has high fixed operating costs: debt service, staffing, insurance, and maintenance do not shrink when parkers leave. That operating leverage means a revenue decline flows almost entirely to net operating income. In the worked example above, a 42% revenue drop produced a 62% NOI decline.
Does falling parking demand make land more valuable? It can. When commuter parking demand falls durably, the land under a garage or lot is often worth more as housing, retail, or mixed use than as parking. That is why more than 100 cities have removed parking minimums and why NAIOP frames excess downtown parking as a conversion opportunity rather than an asset to expand.
Conclusion
Parking real estate was the first property type to tell the truth about remote work. Its income is daily and mark-to-market, so it repriced the collapse in commuting the quarter it happened, while office rent kept reporting a pre-2020 world until leases expired. The two assets faced the same shock. Parking just marked it first.
For the operator, the lesson runs in both directions. On the income side, parking is a live sensor for how a location is actually used, and its high fixed-cost base means demand swings hit NOI with force, as the 62% drop in the worked example shows. On the land side, early repricing exposed decades of overbuilt parking as a low-value use sitting on parcels that zoning reform is now freeing for housing and mixed use. Underwriting parking as a stable, fixed contributor was the error. It is the most demand-sensitive line in the building, and it moved first because it had to.