In both urban and suburban contexts, underground, shared parking can stimulate greater densities using lower actual parking ratios but achieving higher effective ones. Many developers avoid the cost and complexity of mixing three or more uses vertically in a single structure, which results in essentially single-use buildings with their own internal parking, often underground. That practice leads to construction of more parking spaces to accommodate periods of peak demand—spaces that lie fallow and unoccupied at most other times.

Market pressures usually demand that such parking be physically connected to the uses it serves. In suburban contexts, higher required parking ratios and the predominance of single-use buildings reinforce the inefficiency of building redundant peak parking spaces, consuming more land and increasing costs, which together preclude effective horizontal mixed uses.

Urban and Suburban Parking Inefficiencies

In cities with smaller blocks, the problem can be particularly acute. Portland, Oregon, for example, owes a great deal of its character to its small 200-by-200-foot (61 by 61 m) city blocks. City planners and urban historians identify the city’s small grid system as a source of the downtown’s vitality. The many corners of small blocks make exceptional retail locations, and the small blocks allow streets to be light and airy, with buildings built at a more human scale. Small blocks can also stimulate development in smaller increments, in line with market demand.

However, the small grid pattern creates significant inefficiencies for the provision of high-density parking and presents barriers to more intensive development of Portland’s downtown.

The 200-foot-wide street grid is less than ideal for above-ground or underground parking. A typical two-sided parking bay designed for perpendicular parking is 60 feet (18.3 m) wide. That width commonly allows 20 to 24 feet (6 to 7.3 m) for a drive aisle—10 to 12 feet (3 to 3.7 m) for each driving lane—and 18 to 20 feet (5.5 to 6 m) depth for each space on either side of the drive aisle. However, the division of 200-foot blocks by the 60-foot (18.3 m) module leaves 20 feet (6.1 m) remaining, creating at least 10 percent inefficiency in the number of spaces that can fit into a given block. When building cores and ramps are inserted, inefficiency is multiplied. Adding these inefficiencies to the high cost of building underground parking creates significant burdens for downtown developers.

A second source of inefficiency arises when a parking garage is used only or primarily by those who live or work in the buildings above. Each type of real estate development—office, hotel, residential, retail, etc.—has a different pattern of parking use and vacancy. Some are predominantly weekday uses or weekend uses; others are evening or overnight uses. Usually office and retail parking is occupied only about eight hours a day; parking for hotel and residential uses is occupied only slightly longer. For about two-thirds of the day—and all weekend for office use—an investment of up to $50,000 per parking space lies fallow.

Convergence of Planners and Developers

The high cost of developing parking is a difficult issue for city planners and real estate developers. Planners aim to reduce the amount of parking downtown to encourage greater use of transit. However, they often fail to realize that developers share the objective of reducing the amount of parking they need to develop because parking fees do not fully amortize the cost of constructed parking. Developers make money on the units and commercial space the parking supports rather than on the parking itself. More efficient use and occupancy of less parking is in the interest of both developers and planners.

Even if the presence of faster and more frequent transit service can reduce vehicle-miles traveled, any car owned by a downtown resident must have a parking space, regardless of how much the resident drives. So any approach that reduces effective capital costs and increases occupied parking density should be a shared public and private policy objective.

One of the best ways to reduce the inefficiency of a small urban grid pattern, or the large land areas required to park at higher ratios in the suburbs, is to develop multiblock underground parking, which leads to a higher density of mixed uses, making urban and suburban downtowns more livable and prosperous. Many sound reasons exist to adopt public policies and strategies to accomplish this result.

Shared Parking. Parking built under multiple blocks creates a common pool of shared spaces that can directly serve and provide access to any building built above it and, to a lesser extent, commercial development on neighboring blocks. Residents of condominium or apartment structures generally have a strong preference to buy or rent only where they can move groceries and belongings directly into an elevator serving the building in which they live. The same is true, to a lesser degree, for people regarding the buildings where they work or that they frequently visit. This is particularly true in rainy, hot, or cold climates.

Mixing uses vertically in a high-rise building creates such problems as accommodating demand for different floor plate sizes, integrating plumbing chases through several different uses, separating entrances and lobbies, and even finding lenders willing to finance more complicated and expensive mixed-use structures. However, this can be simplified when the mix of uses is horizontal rather than vertical. An active mixed-use environment can be created with fewer design difficulties, cost premiums, and financing challenges.

Shared parking reduces construction costs for a mixed-use building because the same parking spaces can serve different uses at different times. For example, a mix of office and hotel uses is ideal because peak demand for daytime office parking is nearly a mirror image of that for evening hotel parking, allowing significantly fewer spaces to accommodate both uses.

While design, finance, and construction parameters make building a single hotel/office property difficult, it is very feasible to provide a horizontal mixture of a hotel tower and an office tower above a common parking pool. Parking demand for one tower expands as the other contracts. Likewise, residential, retail, restaurant, and entertainment uses have separate parking demand curves and design parameters that can complement one another.

Space Efficiencies. Multiple-block underground parking offers major efficiencies in floor plate size, layouts, ramp design, and retaining wall length and costs, among other elements. For example, a single 200-by-200-foot block (61 by 61 m) underground parking structure will need 800 linear feet (244 m) of retaining walls; four single blocks will require 3,200 linear feet (975 m) of retaining walls. But one four-block underground parking structure with a 50-foot (15 m) right-of-way of streets above will require only 1,800 linear feet (549 m) of retaining walls—44 percent less. Moreover, the multi­block structure will create 202,500 square feet (18,813 sq m) of parking versus only 160,000 square feet (14,865 sq m) in the four single blocks—27 percent more space.

Floor Plate Efficiencies. The impact of the increased space is magnified because it allows much greater efficiency in parking layouts. The most efficient layout for a single 200-by-200-foot (61 by 61 m) block is double-loaded diagonal parking at an angle of 50 degrees for 8.5-foot-wide (2.6 m) parking spaces in bays that are 50 feet (15 m) wide. A parking bay is measured from the tip of one parking space to the tip of the one opposite it, including the drive aisle.

This layout produces a theoretical maximum of 144 spaces per block, excluding any elevator cores or other service functions. Four separate blocks therefore yield at most 576 spaces. But in a multiblock structure, the same four blocks could yield a theoretical maximum of 818 spaces of the same dimensions—an additional 242 spaces, or a 42 percent increase in total parking spaces. The average space per parking stall, including the drive-aisle circulation but excluding ramp inefficiencies, declines to 248 from 278 square feet (to 23 from 25.8 sq m), a nearly 11 percent reduction.

Moreover, fewer spaces are lost to building cores because the spaces under the streets and sidewalks have no such cores penetrating them. In addition, the larger floor plate minimizes the total depth that needs to be excavated for a given number of spaces, as well as the need for multiple ramps and two-way drive aisles that reduce the number of parking spaces for a given square footage. Fewer levels and ramps mean higher efficiencies and lower costs.

Construction and Operation Costs. Excavation of one larger site is significantly more efficient than that of multiple smaller sites, in part because larger earth-moving equipment can be used. Moreover, the excavation need not be as deep to produce an equal or greater number of spaces. Because the need for retaining walls is reduced by 44 percent, that cost is reduced, and because significantly less space is needed per stall, the cost of concrete floors and ceilings is reduced proportionately.

Also, fewer attendants are needed to operate a multiblock structure; entrances and exits can more easily be automated; and the costs of installing gates, pay stations, wayfinding systems, and other technology are spread over more spaces.

Use Density. Increasing the density of development—a major public planning goal in both downtowns and suburbs—can be enhanced by multiblock parking garages. Traditionally, planners have criticized the automobile and the amount of space dedicated to parking and streets, but having denser, maximally occupied parking is the best way to support higher-density uses in downtown buildings. No developer wants to build more expensive and unprofitable structured parking than is absolutely necessary to capture a market and satisfy lenders and equity investors.

Multiblock underground structures and shared parking systems optimize parking investment that can support more hotel rooms and condominiums, as well as office, retail, restaurant, and entertainment space. Accommodating the additional density helps the developer increase profits, helps cities enlarge their tax bases, and meets the public objective of greater density, creating more activity on the streets.

Housing Affordability. Multiblock underground parking structures may also be used to reduce the prices of condominiums built above them. Rather than sell parking spaces with or separately from the condo, if the developer retains ownership of the parking but conveys a deeded option to rent one space per unit, buyers can choose whether to exercise that rental option without losing the right to convey the option to the next resale buyer who may or may not own a car.

Either way, the developer can remove the capital cost of the space from the sale of the condominium apartment, which might lower its price by up to $50,000 in current urban markets. This deeded option to rent can be far superior to selling spaces separately because otherwise, if the first buyer chooses not to purchase a space, that owner’s resale is limited to the smaller segment of potential buyers who do not own a car, which will reduce the resale price.

The effect on housing affordability is even greater. Because buyers will not need to purchase parking spaces to protect their resale prices, more buyers will choose to forgo the capital expenditure of buying parking spaces, as well as the expense of owning and operating a car—estimated at almost $9,000 per year in 2012 by the American Automobile Association. Adding rental of a parking space of about $165 per month raises the cost by about $2,000. Adding that $11,000 per year ($920 per month) to the home price, at a 4.0 percent interest rate on a 30-year loan, adds the equivalent of about $193,000 that could go to the mortgage instead of car ownership.

In the event that a buyer does not exercise the deeded parking option, the developer can rent that space to another resident who wants a second parking space, subject only to termination in the event that the underlying option owner exercises it later or conveys it to a future resale buyer. The net effect is to increase the effective parking ratio for the buyers and to create a new income stream for the developer. Some developers may choose to sell the parking structure for a bondlike return supported by the built-in rental demand from the condominium and other mixed-use components of the project.

Lower-Cost Public Service. Another alternative is for the public sector, which owns the land under streets and sidewalks, to take ownership of the multiblock underground parking structure and manage it as a shared parking pool for all the uses above it as well as for the public, subject to absolute deeded rental options to the residents, owners, and tenants of the buildings. The developer would be compensated by a predetermined purchase price that covers projected costs but shifts cost overruns to the developer.

One virtue of this scheme is that if the economics of private ownership and management are tight, the lower carrying costs of the public sector make the arrangement more feasible, and the public sector can implement policies to ensure that as many housing units as possible above the parking remain more affordable.

Transit Incentive. Counterintuitively, the more parking that is built in multiblock underground parking structures, the more the need to use private automobiles will be reduced. One reason is that a shared pool of parking spaces that supports higher-density mixed uses above those spaces reduces the need for residents to travel elsewhere.

Also, because buyers will not need to purchase parking spaces to protect their future resale prices, more will choose to forgo the expense of buying or renting parking spaces and of owning and operating a car, giving them an economic incentive to switch to transit use.

By entering into public/private partnerships to develop more multiblock underground parking structures, the public sector will be advancing two objectives—building high-density environments that are less car dependent and more transit oriented, and creating incentives to use transit.

Leveraging Car Sharing. Another effect of multiblock underground parking structures is the increased use of car-sharing programs like Zipcar and Car2Go. Initially, car-sharing companies thought their programs would get the most use from residents of outlying residential complexes. Instead, use has been highest for cars stationed at office complexes, where they replace fleet cars for companies and government agencies. That leaves office tenants freer to use transit for commuting, secure in the knowledge that a car is available for daytime trips needed for company or personal business.

Shared cars stationed at office buildings are not likely to be used on evenings or weekends. But if such shared cars are stationed in multiblock parking structures under many residential units, hotel rooms, and offices, the cars are more accessible and likely to be used on evenings or weekends. Because a single shared car may support about six users, total parking space efficiency can be greatly enhanced. And because shared use of the space will give multiblock underground parking structures higher occupancy, fewer new single-use parking structures will be needed.

Public Policies

Because the city owns the streets, it is a participant in any multiblock underground parking structure. No developer controls as much parking as the city, nor could a developer undertake multiblock parking without city involvement. Parking is a public function, just like transportation, and should enjoy tax exemption as a public use.

In the case of Portland’s Brewery Blocks (see sidebar), the Portland Development Commission in 2001 loaned $6 million to the developer at 8 percent interest for ten years to guarantee that at least 400 of the minimum 1,300 parking spaces would be open to the public and priced at the same rate as city-owned SmartPark garages for ten years. The agency also granted the developer $2 million to pay for utility relocation related to the parking structure and ornamental streetlights and sidewalk extensions above to enhance the blocks. Therefore, there are both precedents and reasons for Portland to adopt a clear policy fostering the creation of more such structures and to use its leverage to enhance public development objectives.

Among the ways any city can use that leverage are the following:

  • Negotiate public or joint ownership of multiblock underground parking structures. Public ownership can be desirable for private developers because at current capital costs and market rental rates, underground parking is rarely profitable. However, with lower-cost tax-exempt financing, carrying costs, and operating expenses, the public parking garage can often operate at a break-even point.
  • Use public leasing phased with new multiblock parking projects to support mixed uses. Developers often find it difficult to build multiblock underground parking garages when their projects are developed in phases because they need to carry additional parking costs before development of subsequent phases. With its lower carrying costs, and other public parking revenues, Portland, for example, could lease spaces in a new multiblock garage for its SmartPark pool to encourage development of efficient multiblock structures.
  • Provide property tax abatements for jointly owned multiblock underground structures. To encourage maximal private financial participation in jointly owned multiblock undergrpound structures, the city could provide property tax abatements on the private ownership share. Together with the exclusion of underground parking from FAR limits, tax exemption or abatement can be an important incentive.
  • Promote optimal mixed-use development. Though development programs must always be sensitive to market demand, the city could refine shared-parking analyses to test optimal mixtures of uses to achieve the highest density of uses with the fewest parking spaces. The city could negotiate these optimal mixes when considering leasing subsurface rights under its streets.
  • Ensure that spaces are rented, not sold or assigned. Cities should not participate in any multiblock structure in which spaces are to be sold. Sale or assignment of such spaces permanently removes them from the shared parking pool, thereby increasing vacancy anytime a resident uses his or her car, which is antithetical to public objectives for maximal occupancy supporting optimal mixed-use density.

Multiblock underground parking can be one of the most important tools a city can use to develop its downtown and other urban centers to create the highest density of the optimal mixture of urban uses. Properly implemented, multiblock underground parking can bring about a higher tax base for the city, more affordable housing for its residents, more urban places for its visitors, and more profits for its long-term investors.