Enthusiasm for the Curb: Getting Reduced Parking Minimums Right

One of the important live issues in the current Zoning Ordinance rewrite is the question of off-street parking minimums for residential development.  At present, city code requires new developments to provide a certain number of parking spaces per residential unit.  For example, for each 1-2 bedroom unit, one off-street parking space is required.  The Planning Commission and City Council have suggested that they would like to do away with residential parking minimums in the new code.  Their reasoning is that the provision of parking raises the cost of residential development, effectively forcing residents to pay for parking whether they want it or not.  Further, if residents are forced to pay for parking, they are more likely to choose car ownership over other forms of mobility, since the parking has no “marginal cost” to them. 

Some members and commissioners have advocated for the idea that “the market should decide how much parking is required.”  It may surprise you to hear that we agree with them.  Where we possibly have a different view, at least in emphasis if not in actual direction, is when we consider the pre-requisites for a truly functional market in parking.  If on-street parking is “free”, the market for parking remains highly distorted.   A developer will choose to externalize parking provision to the street.  The result will be street parking in short supply and allocation by environmentally wasteful “cruise time”.  To leave the question of off-street parking “to the market” requires including on-street parking in the market.  The dean of parking scholars, Prof. Donald Shoup, likes to say that before a jurisdiction can dispense with off-street parking requirements, it needs an effective on-street parking policy.

How effective it is from a housing affordability standpoint to get rid of parking requirements depends on the cost that attaches to providing parking.  How critical it is get the on-street parking policy exactly right depends on the slack in the current on-street parking supply.  How easy it is to reconcile these two parts of parking policy depends on the viability of alternatives to car ownership.  In this analysis, we look at each of these issues in turn.  Then, based on what we have found, we make some suggestions for a viable parking policy under the new zoning ordinance.

Cost of Parking Provision

Advocates for the reduction or removal of mandatory residential parking minimums argue that housing costs are increased by the cost to a developer or landlord of complying with parking minimums.  They will often cite the statistic (which appears to trace back to a GAO report on cost factors for LIHTC developments, most of which were in much more expensive jurisdictions than Charlottesville) of $40,000-$50,000 per parking space.  Under a parking minimum regime requiring 1.5 spaces per unit (not atypical), a housing unit’s cost of production could wind up $60,000-$75,000 higher than otherwise.  In a place like Charlottesville, where the average price (according to Zillow) of a condo unit is around $275,000, the cost of parking would represent a significant fraction of the cost of the unit!   We have to ask, though, whether such a cost estimate reflects reality in Charlottesville.  The $40K-50K estimate relates to structured parking of the most complex sort – that is, a parking space in a structure such as the basement of large apartment building.  It does not represent the cost of surface parking.  The new Future Land Use Map and Zoning Ordinance focus on encouraging mid-sized infill development.  It is very unlikely that many such projects will need to provide structured parking solutions. 

Surface parking is much less expensive, consisting mainly of land and paving.  Parking lots need approximately 400 square feet per car.  Paving costs between $3-6 per square foot.  The city’s Inclusionary Zoning report uses as its target estimate for land costs a per-square foot value of $8-10.  At that rate, a parking space would cost somewhere in the neighborhood of $6,000-7,000.  We believe the IZ report underestimates land cost by a factor of two.  Using our estimate, based on actual vacant land and teardown transactions in Cville, the cost per surface parking space would be more like $10,000.  Even this may prove an overestimate.  Land area is not a commodity that is fungible between parcels.  Developers face a “suitcase problem” in that they need to work within the limits of existing parcel boundaries and subdivision/combination possibilities.  Most developments in Cville have lot coverage ratios of well below 50%.  Therefore, surface parking may be a use for land that is “wasted” in the solution to the suitcase problem.  In such a situation, the true marginal cost of a parking space would only consist of paving, amounting to more like $1,500-2,000 per space.  In addition, a landlord or condominium association would face some ongoing maintenance tasks (stormwater fees in the case of use of impervious paving materials, plowing, cleaning, repaving from time to time), but such costs would not rise to particularly significant levels relative to the cost of unit construction or maintenance. 

We can also use data on the price of monthly parking rentals in the city to back into the likely cost of providing off-street parking.  A search of available monthly parking spaces in the downtown Cville area comes up with a range of between $95-$150/mo.  Using a capitalization rate of 7%, that implies a “sale value” in the range of $16,000 – $26,000.  We should consider this an extreme upper bound, inasmuch as these spaces are in the part of town with the most expensive land, some of these spaces reside in parking structures as opposed to surface lots, landlords will need to factor in a certain degree of vacancy, the ongoing cost of monitoring and maintenance are much higher for a pay-lot than for residential parking, and the lot-owner may face the opportunity cost of using the spaces for higher-priced per-hour or per-day parking options.

While we agree that residential parking minimums impose some cost, it is important not to overstate the potential savings derivable from their removal. One of our complaints about the entire CP/Zoning-Rewrite process has been the tendency of advocates to exaggerate the practical benefits of their preferred policy changes — we half-jokingly call it the “One Weird Trick!” school of policy design.  Policy-makers need to compare a realistic assessment of benefits to the costs of a policy such as ending parking minimums.  The removal of parking minimums will not likely make a highly meaningful dent in cost of housing unit production.  Multifamily housing ex-land production cost will amount to at least $200/square foot.  For a 2-bedroom 800-square foot apartment, waiving the requirement to provide one off-street parking space might save a few thousand dollars, but most likely less than $10/square foot.   It’s not nothing, it’s very likely worth pursuing, but it is hardly a panacea.

However, we very much agree that there may be collateral benefits to reduced parking minimums in terms of shifting norms of car ownership and public transportation use.  To be sure, policy-makers should not ignore these potential benefits, even though their quantification is more difficult. 

On-Street Parking Slack

Car-owning households deprived of an off-street parking alternative, or facing a costly one, will look to the neighborhood curbs for parking spaces.  How disruptive this will be depends on the level of slack in on-street parking availability relative to the new demand from marginal housing units without off-street parking.  Below, we share the results of an analysis of available on-street parking, estimated current on-street parking demand, and potential marginal demand from development under the new Zoning Ordinance (using parameters from the FLUM).  The analysis was performed at the street-block level and calculates the percentage of new development entitlement that can be built without off-street parking before on-street parking becomes saturated.  It covers only streets and blocks designated principally Medium-Intensity Residential and General Residential. We use current Census numbers on car ownership for projecting the behavior of new households.  Technical details are available here.  At a headline level, we find that for large parts of the city, it would take rather extreme amounts of development to saturate on-street parking.  However, for many of the areas that we believe would prove most desirable for infill development, as little as a single “medium-intensity” 12-unit development would generate enough new cars to saturate the block.   On the bright side, the blocks nearer to saturation also feature higher-than-average public transit availability and walkability (as measured by WalkScore and TransitScore).  On the downside, these blocks mainly sit in census tracts where car ownership per household already lower than other GR/MIR-heavy areas, so households may not have much room to further reduce car ownership.

Reconciling Minimums and On-Street Management

That many blocks in the city have on-street parking already fairly close to saturation means that in the absence of parking minimums, allowing development without triggering on-street parking difficulties will imply some combination of supply creation and demand destruction.  That’s what “leaving parking to the market means” under a proper understanding a full market for parking, one that encompasses both on-street and off-street parking.  Landlords and developers can charge directly for parking and provide parking according to the economic attractiveness of the proposition; the city must implement an on-street parking policy that extends market principles to on-street parking.  Residents will bear the cost of parking spaces they use.  In essence, between landlords and the city, we will have a sort of Pigouvian tax on car ownership.  Some of the externality of car ownership will be internalized.  The “tax” will provide an incentive both for supply creation (landlords building off-street parking) and demand destruction (residents reducing car ownership to avoid the cost of parking).   If the adjustment comes entirely through supply creation, then the city will find itself in a housing-cost position not far off from if it had left minimums intact.   The hope is that the policy would lead to some demand destruction.  The extent of the reduction in car ownership depends on how difficult it is for residents to respond – for, say, two-adult households to have one car instead of two, or for households to forego car ownership entirely.  That in turn depends on the feasibility of living without a car – not merely foreswearing a daily car-commute, but rather eliminating the car entirely.  It is encouraging that the near-saturation blocks tend to be better serviced by public transportation and are more walkable than average.  Blocks in the lowest quartile of slack had a median TransitScore of 51, while blocks in highest quartile of slack had a median TransitScore of 43.  The results for WalkScore proved even more dramatic: 66 and 38 respectively.   Yet, we have to recognize that even a TransitScore of 51 does not correlate to widespread abandonment of car ownership by households.   Cities with among the highest TransitScores in the nation come in at 70 or above.  Even in these, zero-car households represent a small minority.  Boston has a TransitScore of 74 and 33% of households go without any vehicle.  Philadelphia has a TransitScore of 68 and around 28% of households are car-free.   Portland, at 50, has a TransitScore in line with Charlottesville’s “near-saturation” blocks, and 13% of households are car-free.  Charlottesville is already at 11% of households car-free.  For our highest “car-free” census tract, the number is 19%, and many of these are likely student households.  According to ACS data, 75% of Charlottesville residents who work outside the home commute to work by private vehicle. Only 6% take the bus. Bikers are a rounding error. While 15% walk, well over half the walkers live in the three census tracts that bracket UVA Grounds, suggesting limited feasibility of walking in other parts of the city. Without major improvements to transit systems, pedestrian infrastructure, and bike infrastructure, it is difficult to imagine that a large number of the households moving into new development would go entirely without a car.  The interactive map below allows you to select a means of commute (use the layer control in the upper right) and see what percentage of commuters use that means in each Census block group. The only block groups where less than half of commuters go to work by car sit immediately adjacent to UVA or UVA hospital. Proximity of large job centers, rather than housing characteristics, appear to determine the use of alternatives to private vehicles. Midrises in Greenbrier will not be full of public-transit users, at least not until the city is transformed in ways that go beyond housing form.

Still, we think that the city could remove residential parking minimums without causing any big problems (and perhaps reaping meaningful environmental benefits), as long as an appropriate on-street parking management system operates.  The removal of parking minimums may not appreciably lower the cost of housing production or massively change car ownership behavior, but at least it would not lead to on-street parking chaos and traffic-choking, polluting “parking cruising.” 

Finally, in order to encourage the effect of removed mimimums to manifest in lower car ownership rather than merely higher explicit costs for (the same number of) car owners, the code could require “mobility minimums” in place of parking minimums. That is, landlords should be required to provide tools for non-car mobility. For example, code might require landlords to bundle public transit passes with rent, which would create a guaranteed revenue stream for CAT (when fares are reimposed) and make public transit use “zero-marginal-cost” for residents. Likewise, code might require bike lockers or bike rooms, as well as infrastructure for charging and storing micromobility vehicles like scooters.

Ideas for On-Street Parking Management

Charlottesville already has some degree of on-street parking management.  City ordinance calls for required residential parking permits for a few areas in the city.  Mostly these areas lie in close proximity to the University or UVA Hospital.  The aim of the ordinance is to prevent commuting students and UVA employees from monopolizing on-street parking in nearby neighborhoods.  A measure of how much demand there is already for on-street parking in central neighborhoods is that in these small permit-only parking zones the city issues over 2,000 permits per year.   The permits, however, cost a mere $25 per year, so many people may secure them without a genuinely strong need.

An on-street management system strict enough to withstand the influx of a large number of households that will either not have access to off-street parking or only have access to it at market prices will require much higher prices and more extensive coverage.  It might also need to make provision for households that have organized their lives around the free availability of on-street parking and people with disabilities to get free or reduced-price access to on-street parking. 

With this background in mind, we limn the outlines of one possible on-street parking management scheme that could enable the removal or reduction of residential parking minimums.

  • Parking minimums are removed only in those areas where on-street parking management is simultaneously put into effect.
  • An initial phase of removal of parking minimums and imposition of on-street management will take place in areas of the city where both the TransitScore and WalkScore are above 50.  This will ensure that there is a good chance that the policy will lead to changes in car-ownership behavior rather than merely moving the cost of parking provision from implicit to explicit.
  • For the same reason, to be eligible for the waiver of parking minimums, landlords should be required to provide “mobility minimums” instead.
    • Bundle public transportation passes with rent.
    • Provide a minimum number of bike lockers or bike room space per bedroom.
    • Provide micromobility charging infrastructure.
  • Streets under on-street parking management will have at least 80% of each block’s parking frontage reserved for permit-holders.  Remaining parking will be open or metered.
  • Resident permits will be made available to residents partly by per-parcel entitlement and partly by auction.
  • Permits will be tied to a specific vehicle (plate #), to avoid arbitrage. The city could consider requiring that the associated car be registered to an address in the city, which would have the collateral benefit of reducing personal property tax avoidance.
  • Each parcel will get an allotment of nominal-fee resident permits based on parcel street frontage.  One permit will be allocated per 40 feet of street frontage.
  • People with disabilities will be eligible for free permits (assuming they don’t already have a handicapped space in front of their property)
  • The city will calculate the remaining available parking capacity after allocation of free/nominal cost permits, per zone, each zone consisting of a few square blocks.  90% of remaining capacity will be made available by an annual online auction process.  Each housing unit will be able to bid for up to 4 permits. 
  • A small number of permits will be held back for sale (at the clearing price) to residents who move in mid-year.
  • Guest permits will be available for purchase to any resident of a given zone.  Guest permits will be good for one day and will be priced at a multiple of the equivalent per-day clearing price of annual permits.  Think something on the order of $10/day.
  • Revenues from permits and fines will be allocated to public transit, bike and pedestrian infrastructure.

October 2022