Workplace parking policies in an agglomeration: An illustration for Barcelona
Introduction
Employer-paid or highly subsidised parking is a very common practice (Inci, 2015; Small and Verhoef, 2007). In Barcelona, for example, 45% of car trips to work use an employer-provided parking space, which is free for employees in 81% of the cases (ATM, 2015). In non-urban or less densely populated areas, the proportion of employer-provided parking may be much higher. Van Ommeren and Wentink (2012) report that in the Netherlands, 80% of car commuters use employer-provided parking, while, as per Shoup (2005a), 95% of car commuters in the US park free at work. In the US, costs of employer-provided parking represent around 1% of the GDP (Shoup, 2005a).
In this context, a key question is why employers provide their employees with such parking benefits, which are rarely offered to those using other transport modes. This is mainly due to two policy distortions (Evangelinos et al., 2018; Shoup, 2005a; Van Ommeren and Wentink, 2012). First, employer-paid parking is generally exempt from fringe benefit taxation, while wages and most perks are subject to income taxation. Hence, wage increases to compensate the loss of the free parking benefit would be more costly for the firm as it should also incorporate the income tax at marginal rates. This provides incentives to employers not to charge parking costs to their employees. Second, land use regulations may impose minimum parking requirements on office buildings. This generally leads to an excessive supply of parking, i.e. the marginal value of parking is below its resource costs (Cutter and Franco, 2012). Even if these regulations may be reviewed,1 most of the supply of workplace parking in different sectors is still determined by such parking requirements.
Employer-paid parking leads to different types of economic inefficiencies. First there is the distortion in modal choice: car use is made cheaper than it really is. Thus, providing free or low-cost parking to employees clearly exacerbates the problem of urban traffic externalities, such as congestion, air pollution, accidents and noise. Second, too much land and capital resources are dedicated to employer-provided parking (Shoup, 2005b). Van Ommeren and Wentink (2012) use data in the Netherlands to show that exemption from fringe benefit taxation and minimum parking requirements, similar to those applied in the US, would induce, welfare losses of about 10% and 18%, respectively, of parking resource costs.
This paper studies two policy responses to employer-paid parking: a switch to employee-paid parking by increasing the cost of parking perceived by commuters and a workplace parking levy, whereby employers are charged for the number of parking spaces they provide to their employees.
The analysis of these policies requires a general equilibrium approach where workers can choose residential and workplace locations, mode of transport and type of parking. Employers and land developers can choose where to produce and where to develop housing.
In urban general equilibrium modelling, there are several traditions (Proost, Thisse, 2019). One is the use of computable land use models (Anas and Liu, 2007), more recently there is the quantitative spatial economics tradition (Desmet, Rossie-Hansberg, 2013). These two model approaches do not allow an analytical solution and miss the detail of the parking market. We will use the third tradition known as the analytical mono-centric model tradition that allows a simplified representation of land used for parking.
Our research builds upon the stylised model of Brueckner and Franco (2018). They use a ‘two regions’ urban model with road and public transport modes and with employers in the central business district providing surface parking. Our model picks up some of their suggestions. First and foremost, we include agglomeration economies by allowing workers to choose between work in the city centre where agglomeration economies are at work and the suburbs where average product is constant. Second, while Brueckner and Franco (2018) focus mainly on surface parking, our model assumes that off-street parking facilities in the city centre are built underground, which is more representative of dense, European-style cities. Third, to capture the potential parking spillover, we consider on-street parking in residential areas as an alternative to employer-provided parking. Fourth, we introduce crowding discomfort in public transport. The introduction of congestible public transport makes the model more realistic: non-congestible public transport sets an unrealistic upper-bound on the commuting cost (the public transport cost). Our commuting cost is convex in the total number of commuters. Fifth, besides the analysis of a switch to employer-paid parking, we include a workplace parking levy as a second-best pricing instrument for urban traffic externalities.
We illustrate our model with a numerical example for the metropolitan area of Barcelona. This aims to quantify the relative importance of the different effects (i.e. agglomeration economies, land use, transport and parking) in a second-best setting. As one of the densest urban areas in Europe,2 Barcelona is an illustrative example of a compact city with underground parking facilities. Moreover, since 45% of car trips to work use an employer-provided parking space and only 19% of employees pay for parking (ATM, 2015), current workplace parking policies in Barcelona are associated to persistent congestion and air pollution issues, while no road tolling scheme has been implemented as of yet. As such, the metropolitan area of Barcelona is an appropriate context for the analysis of a second-best solution to urban traffic externalities based on workplace parking policies.
Compared to Brueckner and Franco (2018), our model illustration allows us to derive two additional insights on the effects of workplace parking policies. Most importantly, we find that, as we introduce agglomeration economies, a switch to employee-paid parking promotes employment in the city centre and increases average productivity. In our numerical analysis for Barcelona we find that the agglomeration benefits are three times as large as the effects on the transport and land use markets.
Second, while in Brueckner and Franco (2018) the reduced suburbanisation effect of employee-paid parking is only attributed to the availability of surface parking at the workplace, we show that, in cities with underground parking facilities, employee-paid parking also reduces urban sprawl when spillover parking to residential areas is limited. Important to note is that our driver for the increase in number of central residents is a higher generalised cost of commuting (road congestion combined with congestion in public transport) caused by employee-paid parking rather than the generation of more land available for housing in the centre due to the reconversion of surface parking to residential use, as concluded in Brueckner and Franco (2018).
Our paper is structured as follows. Section 2 reviews the literature. Section 3 describes our stylised basic model and its main assumptions. Section 4 derives the social optimum and Section 5 sets the first-best benchmark based on the equilibrium conditions for the parking, transport, labour and housing markets. Section 6 uses comparative statics to analyse the effects of workplace parking policies on travel behaviour, agglomeration economies and residential location. Section 7 deals with the second-best setting of a workplace parking levy when other policy instruments are restricted. Section 8 presents the numerical illustration for the city of Barcelona. Finally, Section 9 concludes.
Section snippets
Literature review
Economic theory suggests that, in the absence of other market distortions, urban traffic externalities can be corrected with marginal cost pricing for both the trip and the parking price. For parking, this includes marginal resource and user time costs (Arnott, 2011). In a second-best setting, however, imperfect road pricing and parking charges should be simultaneously determined so that, when parking fees are raised or all employees have to pay for their parking, road charges can be reduced (
Spatial structure
Following Brueckner and Franco (2018), in our model a city is divided into two regions connected by a road and a public transport link. Both the central city (denoted by subscript c) and the suburbs (denoted by subscript s) have a mixed land use including production and residence (see Fig. 1).
The land supply in the central area is perfectly inelastic, with a fixed land use split between production land and residential land , and a land rent for residence in the centre .5
Social optimum
The optimal allocation from a social planner's perspective is reached when the uniform utility is maximized subject to a resource constraint of the city's economy. The production of the city and the suburb sets the available resources for land and non-land consumption and for transport and parking. The optimal allocation can be studied by the following constrained optimisation of the uniform utility level:
Equilibrium and first-best decentralisation
The social optimum can be decentralised as a competitive equilibrium if sufficient policy instruments are available. The following policy instruments can be potentially applied to reach a first-best decentralisation:
- •
a wage subsidy in the centre
- •
a user fee on public transport per round trip
- •
a road toll per round trip
- •
a policy measure that eliminates employer-paid parking
- •
a workplace parking levy per day for employer-provided parking spaces
- •
an on-street parking fee per day
Analysis of the effects of workplace parking policies
In this section, we use comparative statics, to analyse the effect on travel behaviour, agglomeration economies and residential location of the two analysed policy instruments related to workplace parking: a switch to employee-paid parking and a workplace parking levy. The order of magnitude of these effects will be illustrated numerically in section 8.
Second-best workplace parking levy
Suppose that we cannot count on all the policy instruments required to reach the first-best decentralisation. It seems interesting to analyse if the application of a workplace parking levy could be a second-best policy instrument to deal with road congestion.
Say that, for example, because of political constraints, road tolls cannot be introduced (). In such case, a workplace parking levy () may be regarded as a second-best policy instrument to discourage road use and reduce the road
Numerical illustration for Barcelona
In this section, we illustrate our theoretical model with a numerical example for the city of Barcelona. In this way we can determine the relative importance of the different effects. This is particularly important when we are in a second-best environment because the first best is not always providing the correct policy guidance. We start by outlining the setting and the functional forms of the numerical case, we then present the data and the calibration process and, finally, we discuss the
Concluding remarks
This paper studies the effects of workplace parking policies on the urban economy through a stylised analytical model adapted from Brueckner and Franco (2018) and a numerical illustration for the metropolitan area of Barcelona.
First, as Brueckner and Franco (2018), we find that both switching to employee-paid parking and the application of a workplace parking levy mitigate congestion by shifting commuters from road to public transport. This continues to hold when there is a limited shift from
Author statement
Aleix Pons-Rigat: Conceptualization, Methodology, Software, Formal analysis, Investigation, Writing – original draft, Writing – review & editing. Stef Proost: Conceptualization, Methodology, Writing – original draft, Writing – review & editing. Mateu Turró: Writing – original draft, Writing – review & editing.
Acknowledgement
This work was supported by the Autoritat del Transport Metropolità de l'àrea de Barcelona (ATM).
References (51)
Congestion tolling with agglomeration externalities
J. Urban Econ.
(2007)Spatial competition between parking garages and downtown parking policy
Transport Pol.
(2006)Congestion, agglomeration, and the structure of cities
J. Urban Econ.
(2016)- et al.
Employer-paid parking, mode choice, and suburbanization
J. Urban Econ.
(2018) Agglomeration effects in Europe
Eur. Econ. Rev.
(2002)- et al.
Do parking requirements significantly increase the area dedicated to parking? A test of the effect of parking requirements values in Los Angeles County
Transport. Res. Part A Policy Pract.
(2012) - et al.
Evaluating the impact of a workplace parking levy on local traffic congestion: the case of Nottingham UK
Transport Pol.
(2017) - et al.
Workplace Parking Levies: the answer to funding large scale local transport improvements in the UK?
Res. Transport. Econ.
(2014) - et al.
The economics of crowding in rail transit
J. Urban Econ.
(2017) - et al.
Pricing workplace parking via cash-out: effects on modal choice and implications for transport policy
Transport. Res. Part A Policy Pract
(2018)