Chapter 5. Steering urban development to more sustainable pathways

This chapter discusses policies to control urban sprawl and direct urban development to more sustainable pathways. It describes a number of land-use and transport policy instruments that can be used to achieve these objectives and analyses their potential benefits, but also the challenges arising in their implementation. The land-use policy instruments analysed include urban containment policies (e.g. greenbelts, urban growth boundaries), minimum density restrictions, property taxation, development rights and incentives for private provision of public infrastructure. Road pricing, parking policies, motor fuel taxes and investments in sustainable transport infrastructure are the main transport-related policy instruments considered in this chapter. The chapter also highlights the importance of taking into account interactions between different instruments when it comes to the development of urban policy. Finally, it emphasises the need for an integrated approach to make urban development patterns more sustainable.


5.1. Introduction

In many OECD cities, urban development patterns are environmentally, economically and socially unsustainable. Emissions of greenhouse gases and harmful air pollutants are increasing, income inequality is rising and economic growth is hampered by congestion and high costs of providing public services. As discussed in Chapter 4, in many cases, urban sprawl underlies these problems.

Urban sprawl has important private benefits, which should, however, be weighed against their social costs. Urban sprawl is to some extent the result of individual preferences for more space, comfort, safety, privacy, easier access to natural amenities and an aversion to the noise and pollution usually encountered in densely populated areas. It is important to notice, however, that such private benefits of urban sprawl are generally internalised in market prices, and are thus reflected in property values in suburban areas. In sharp contrast, the external costs of urban sprawl, as presented in Chapter 4, are not incurred by property owners in suburban areas. As market forces fail to take external costs into account, policy intervention is in many cases necessary to achieve desirable environmental, economic and social outcomes.

This chapter discusses policies to mitigate the environmental and economic costs of urban sprawl. The chapter describes an array of land-use and transport policy instruments that can be used to this end and discusses their merits and drawbacks. It also provides a word of caution regarding the possible side-effects of policies to control sprawl. In addition, it explains the critical role of policy interactions, both between sectoral policies, e.g. transport and land-use policy, and between policies implemented by different jurisdictions, e.g. by local authorities and national governments. The chapter also emphasises the need for following an integrated approach to direct urban development to more sustainable pathways.

Drawing direct links between specific dimensions of urban sprawl and policy instruments is usually very complex. Most of the policies analysed in this chapter are likely to affect more than one dimension of the phenomenon. For example, the introduction of road pricing would encourage denser and – probably – less fragmented urban forms, as it would increase the private costs of car use, inducing households to move closer to their jobs and their daily activities. However, policies do not always influence different dimensions of urban sprawl in the same way: in fact, a number of policies can affect different dimensions of the phenomenon in opposite ways. For example, while urban containment policies, such as greenbelts, may increase average population density, they can also increase fragmentation and the share of urban population residing in areas of very low density. The reason for this is that, under specific circumstances, greenbelts can cause leapfrog development in remote areas.

The heterogeneity of the impacts of policy instruments on urban sprawl highlights the usefulness of the approach developed in this report. The multidimensional definition and the indicators of sprawl presented in the previous chapters enable a deeper understanding of the impact of each policy on urban form and a more informed assessment of its effectiveness in mitigating the environmental and economic problems caused by urban sprawl. The discussion of specific policy instruments later in the chapter draws to the extent possible on the potential impacts of each instrument on different dimensions of sprawl.

Providing policy recommendations that retain their validity across highly heterogeneous urban settings is particularly challenging. Each city is a unique socioeconomic system, where only a small set of policy mixes can be economically efficient, environmentally effective and socially acceptable. However, different urban areas often face similar environmental and economic challenges. Provided that the behaviour of individuals and firms is guided by the same economic principles in all urban areas, it is possible to identify policy options which are likely to help meeting these challenges in different urban contexts.

Urban sprawl may be manifested at the city level, but its causes and consequences are not contained by the urban limits. Greenhouse gas emissions stemming from sprawling urban forms contribute to climate change and pressures on local finance, which often lead to additional burden for national or regional budgets. Local policy action to address the negative effects of sprawl, e.g. at the municipality or county level, can be undermined by the action of neighbouring local authorities which may engage in policies favouring sprawl or refrain from policy action for their own interests. These challenges call for action from national and regional governments to curb the consequences of urban sprawl.

The policy directions discussed in this chapter are in alignment with the targets of the United Nations’ SDG 11, which aims at promoting inclusive, safe, resilient and sustainable cities. Among others, these targets pertain to: i) the reduction of the adverse per capita environmental impact of cities; ii) the support of positive economic, social and environmental links between urban, periurban and rural areas by strengthening national and regional development planning; and iii) access to safe, affordable, accessible and sustainable transport systems for all by 2030 (United Nations, 2015).

The structure of the chapter is as follows.  5.2 discusses concrete policy actions to control urban sprawl and mitigate its consequences.  5.3 highlights the importance of considering interactions between policy instruments and following an integrated approach to designing and implementing urban policies.  5.4 concludes.

5.2. Taking steps to reverse unsustainable urban development patterns

Developing the right policies to reverse the unsustainable trends often observed in urban development is a very challenging task. This section does not attempt to provide city-specific guidance on promoting sustainable urban development. Instead, it discusses a set of policy actions which can help achieve sustainable urban development in the long run in various urban settings. Policy actions are categorised according to the main objectives they aim to achieve: promoting socially desirable levels of population density; or mitigating the adverse environmental and economic effects of sprawl-induced car use. The proposed policy actions are not equally relevant for each urban area: their relevance and potential effectiveness depends inter alia on geomorphological factors, population growth and the governance system. Policy instruments which are powerful in specific contexts may have negligible effects or even lead to undesirable consequences in different contexts. The section also puts forward some general directions for the development of policies to control urban sprawl and address its consequences.

Promoting socially desirable levels of urban density and fragmentation

Urban sprawl is manifested in low-density and fragmented development, which may have serious environmental and economic implications. Car dependency and accompanying emissions from car use, loss of open space, and high costs of providing public services, such as waste management, public transport, water supply and wastewater treatment, are some of these implications, which have been discussed in detail in Chapter 4.

High density and contiguity are important characteristics of sustainable urban development, but it is also important to mention some caveats. First, high density should not be considered as a synonym for high average density. Certain cities may be overall dense, while also containing large artificial areas of very low density. Such a development pattern will rarely be sustainable. Thus, policies promoting densification may also be very relevant in the case of these cities. Second, while high-density development can help prevent multiple environmental pressures, it also increases population exposure to a given level of air pollution, noise and natural disaster risks. The external costs of air emissions and noise increase with population density and, thus, corrective policies (e.g. taxes or pollution-control regulation) should be more stringent in high-density areas. At the same time, high-density development in areas vulnerable to natural disasters (e.g. earthquakes, floods, tsunamis) entails higher risks for residents which should be reflected in higher property insurance premiums and/or stricter building codes. Third, infill development policies that incentivise the connection of existing urban fragments should account for the impact such development may have on biodiversity. For example, special care is required when designing policies connecting fragments in suburban areas, as such development may block corridors that are critical for local biodiversity. Infill development should take place in a way that does not induce further natural habitat fragmentation. In case urban development leads to further fragmentation of natural habitats, policy instruments aimed at offsetting or compensating for biodiversity losses need to be considered.

Traditionally, spatial planning and land-use policy has overwhelmingly relied on regulatory instruments to control urban sprawl and mitigate its consequences. Regulatory instruments are used to determine where urban development will take place, where specific land uses will be allowed, and in many cases how urban population will be distributed across space. Regulatory instruments vary inter alia in terms of stringency, spatial and temporal control, and links with public infrastructure needs. Despite their widespread use, it is not clear whether, and under what conditions, these instruments eventually result in denser or less fragmented cities.

Land-use policy has leveraged market-based instruments, including taxes on immovable property, and purchase and transfer of development rights, to a much lesser extent to promote socially desirable levels of density and fragmentation. Immovable property taxes are ubiquitous in OECD countries, but they are mostly viewed only as a revenue raising mechanism. By contrast, their potential to encourage more sustainable land-use patterns largely remains to be explored.

Reform urban containment policies

Urban containment policies are perhaps the most direct form of policy intervention to control urban expansion. They essentially set boundaries to urban development, motivated by intentions to protect surrounding forestland and farmland, reduce the costs of public service provision and induce development in vacant land within the contained area (infill development). Greenbelts and urban growth boundaries are the most popular urban containment policies. Box 5.1 defines these two types of policies and describes their main characteristics.

Box 5.1. Greenbelts and urban growth boundaries

A greenbelt is a band of natural open space surrounding an urban area, which is designated to function as a permanent boundary to the expansion of the city. In contrast to other land-use policies which are introduced by local authorities, greenbelt policies are often developed at the national or regional level. They are created through public acquisition of open space, purchase of development rights, or through regulation of private property (Bengston, Fletcher, and Nelson, 2004; Bengston and Youn, 2006). Greenbelts are common in the UK, Germany and Korea, while they have also been implemented in urban areas of other OECD countries, including Australia, Canada and the US (Amati, 2016; Bae and Jun, 2003; Lee and Linneman, 1998; Siedentop, Fina, and Krehl, 2016; Taylor, Paine, and FitzGibbon, 1995; Vyn, 2012).

Urban growth boundaries are limits to urban expansion set around the urban area, with the purpose of protecting its surroundings from urban encroachment. In contrast to greenbelts, urban growth boundaries are not intended to be permanent; instead, they are periodically evaluated and, if necessary, expanded. Urban growth boundaries are generally implemented by the use of regulatory policy instruments. Likewise, urban service boundaries designate the area within which public services, such as water supply and sewerage, are offered. Provision of these services outside of the boundaries is not permitted. Urban service boundaries are sometimes used in combination with adequate public facilities ordinances to prevent urban expansion beyond the designated area (Bengston, Fletcher, and Nelson, 2004).

Source: Amati, 2016; Bae and Jun, 2003; Bengston, Fletcher, and Nelson, 2004; Bengston and Youn, 2006; Lee and Linneman, 1998; Siedentop, Fina, and Krehl, 2016; Taylor, Paine, and FitzGibbon, 1995; Vyn, 2012.

Urban containment policies are generally very effective in achieving their objectives, i.e. protecting designated open space and inducing infill development. Infill development results in a denser and possibly less fragmented contained area. However, these features do not extend to the rest of the urban area (Bengston and Youn, 2006; Gennaio, Hersperger, and Bürgi, 2009). That is, urban containment policies are known to give rise to leapfrog development beyond the contained area, which often implies an overall more fragmented, and probably less dense, urban area. In turn, these characteristics induce longer commuting distances (Amati and Yokohari, 2006; Bae and Jun, 2003; Bengston and Youn, 2006; Jun, 2004; Vyn, 2012), increase emissions from mobile sources and raise the costs of public transport provision (see also Chapter 4). The above effects are further fuelled by the fact that the modal split depends on the commuting distance. Thus, not only does leapfrog development increase distances between key points of economic activity, but it also affects positively the probability that these will be covered with private vehicles. Thus, the magnitude of the side-effects associated with urban containment policies should be carefully evaluated.

A key determinant of the magnitude of the side-effects of urban containment policies is the amount of developable land remaining in the inbound area, i.e. the area between existing developments and the greenbelt or the growth boundary (see e.g. Siedentop, Fina and Krehl, 2016). A low potential for new urban development in the inbound area may lead to a significant reduction in the elasticity of housing supply, which will, in turn, trigger increases in housing prices (see also Blöchliger et al., 2017; Caldera Sánchez and Johansson, 2011). The higher the increase in housing prices is, the stronger the incentives for leapfrog development are. In contrast, when the containment policy leaves sufficient space for urban development in the inbound area, the extent of leapfrog development is likely to be limited. In that case, however, urban development in the contained area might not be as dense as desired (Siedentop, Fina and Krehl, 2016).

An important determinant of the effectiveness of urban containment policies is their flexibility, especially in terms of allowing for periodic reviews and, if necessary, reforms. For example, a major criticism of greenbelts concerns their permanent character. This lack of flexibility may undermine the effectiveness of the policy, especially when it is implemented in urban areas facing rapid population and economic growth. Increases in housing prices, and leapfrog development in areas beyond the greenbelt are reinforced by the stringency of the policy and significant population pressures that may be present (Amati and Yokohari, 2006; Bae and Jun, 2003; Lee and Linneman, 1998).1

The effectiveness of urban containment policies is also influenced by the governance system. In a decentralised system, where the various jurisdictions have the power to administer their own spatial planning and land-use policies, leapfrogging can be intense. This is because the jurisdictions whose fiscal dominion lies beyond the contained area may attempt to attract residents and businesses by implementing laxer policy frameworks than jurisdictions within the contained area. In centralised systems, it is much easier to ensure that individuals are not provided any perverse incentives to reside beyond the contained area. This practice, however, may reduce the price elasticity of housing supply (as developers will be allowed to build in very few locations) and result in higher housing prices and lower housing affordability (Ball et al., 2014; Bengston and Youn, 2006; Blöchliger et al., 2017).

Leapfrogging would be much less of an environmental concern if firms were closely following the movement of their labour pool. In that case, leapfrog developments would eventually be transformed to new urban centres, a polycentric urban structure would emerge and commuting would be largely limited within each centre. However, real-world examples of urban containment policies, such as Seoul’s greenbelt, show that in the absence of strong relocation incentives, firms may not be willing to forego the benefits of agglomeration economies in the contained area (see e.g. Bae and Jun, 2003).

Apart from their potential side-effects on urban expansion, external environmental and congestion costs, and housing prices, urban containment policies have also been heavily criticised for creating winners and losers, who see the values of their properties change (in some cases, dramatically). Losers are frequently not compensated for the fall in their property values.2 In the absence of mechanisms to compensate the losers, the policy will be perceived as unjust and pressures for policy change may be triggered. It is, thus, important to accompany urban containment policies with instruments providing some form of compensation to the losers from the policy intervention.

In summary, urban containment policies are likely to have side-effects, manifested in leapfrog development, excessive commuting and increased housing prices, which can jeopardise their effectiveness in controlling urban sprawl and tackling environmental problems. The magnitude of these side-effects depends inter alia on: i) the level of centralisation of the governance system, ii) the stringency of the policy (amount of developable land in the contained area), and iii) the stringency of other land-use and transport policies implemented in the urban area. To some extent, such side-effects can be prevented by introducing a small level of flexibility in stringent urban containment policies, and, most important, by coupling them with laxer density limits.

Relax maximum density restrictions

Maximum density restrictions are common in urban areas of OECD countries and beyond. Most often taking the form of building height restrictions, i.e. limits on building floor-to-area ratios, these policies inevitably hamper densification (Bertaud and Brueckner, 2005; Brueckner and Sridhar, 2012). Despite being relevant in specific locations, such as close to airports, historic buildings or other elements of cultural heritage, stringent maximum density regulations are in many cases unwarranted. Such regulations may lead to low average density levels and, when being overly stringent, are likely to increase the share of urban footprint hosting very low density levels, with important environmental implications (e.g. in emissions) presented in detail in Chapter 4.

Policy makers should be aware of the negative interactions between maximum density regulations and urban containment policies. When the restrictions imposed by these regulatory instruments are too stringent, leapfrog development in remote areas and soaring housing prices in the contained area are very likely outcomes. One of the few cases that this combination of stringent instruments may not have undesirable consequences is when areas in the vicinity of the contained area are zoned for high-density development. Otherwise, actions to relax policy stringency are warranted in order to direct urban development to a more sustainable pathway with lower levels of fragmentation, a more elastic supply of residential floor space and more affordable housing. Such actions may be justified even in the absence of urban containment policies. Further to upzoning (i.e. allowing higher densities than the ones permitted before), it may also be worth considering introducing minimum density restrictions in areas where densification is mostly needed (see also Bengston, Fletcher, and Nelson, 2004; Silva and Acheampong, 2015). It should be noted that relaxing quantity restrictions in housing markets may prove to be a policy challenge per se. Current homeowners often have strong incentives to lobby in favour of building height restrictions, as well as other types of regulatory mechanisms that reduce the supply of floor space (Fischel, 2001; Schuetz, 2009), as such regulations may increase the housing prices of the existing dwelling stock (e.g. Ihlanfeldt, 2007).

Reform property taxation to better reflect the social cost of certain urban development patterns

Property taxes are the most commonly used market-based instrument applied to land use. A property tax is a recurrent ad valorem tax on immovable property, usually levied by the local authority of the jurisdiction where the property is located. In practice, the primary objective of property taxation has been to raise public revenues for the financing of local services (see e.g. Brandt, 2014), rather than to address the negative externalities of urban development. For example, in Italy, property taxes accounted for about 48% of total municipal revenue in 2010 (Ermini and Santolini, 2017).

In most countries, property taxes are levied on both land and land improvements, i.e. buildings and other structures (Bird and Slack, 2004). This form of taxing property has been argued to be a combination of one of the least distortionary taxes – the tax on land value – and one of the most distortionary ones, the tax on land improvements (Vickrey, 1996). Box 5.2 discusses the merits of land value taxes, but also the possible challenges in their implementation.

Box 5.2. Land value taxes

A land value tax is essentially a tax on an economic rent and, thus, it does not discourage any desirable economic activity, such as labour supply or investment. The amount of land is fixed and land supply is fully inelastic – at least in the absence of regulatory land-use policy. This entails that after-tax prices of land would be equal to pre-tax prices and the tax burden would be fully incurred by landowners. The tax burden on landowners does not distort economic efficiency and has long been advocated as a fair outcome (Mirrlees et al., 2011). Land rents reflect windfall gains made by landowners as a result of community efforts (e.g. construction of public infrastructure) and land-use regulation; they do not stem from effortful activity from their side (George, 1879).

Land value taxes incentivise the allocation of land to its most valuable use, which will often be residential or commercial development. They will have desirable effects when it comes to unused land located close to existing infrastructure or in urban cores, as they will induce development where it is most needed. However, land value taxes could also induce development in other areas of high value, such as the ones close to ecologically sensitive areas, including forests and coastlines (Brandt, 2014). Thus, land value taxes would not necessarily lead to less sprawled cities or lower environmental pressures from development. This is why they should be used in combination with some form of regulatory or market-based policy instrument, which would discourage development in ecologically sensitive areas (e.g. zoning, or preferential tax treatment of agricultural and forestry use).

Despite their appeal on economic efficiency and equity grounds, land value taxes have rarely been used in reality. Administrative complexity in assessing land values and political economy reasons explain to a large extent the lack of real-world applications. One of the most important challenges in their implementation is the difficulty in disentangling the value of the land from the value of the improvements made on it (Mirrlees et al., 2011). However, obstacles to the implementation of land value taxes are not always insurmountable and economic arguments for their implementation are strong. Hence, it is important that they are not left completely unconsidered by policy makers.

Source: Brandt, 2014; George, 1879; Mirrlees et al., 2011.

The direction of the effect of property taxes on urban density is highly uncertain, even on theoretical grounds. Most property tax systems apply the same rates to the value of land and buildings. As discussed in Chapter 4, higher property taxes may result, ceteris paribus, in lower land prices, especially when land supply tends to be inelastic. In that case, an increase in the property tax rate may induce development with a lower floor-to-area ratio and a larger footprint, resulting in a less dense urban structure. On the other hand, higher property taxes increase house prices and, therefore, reduce the demand for residential space. Smaller properties imply a smaller urban footprint, more densely developed areas and less land being devoted to urban use. The direction of the effect of single-rate property taxes on urban density is, therefore, theoretically unclear (Brueckner and Kim, 2003).3 The relatively thin empirical literature investigating the effect of property taxation on urban sprawl does not lead to conclusive evidence either.4

Urban areas usually encompass multiple jurisdictions (e.g. municipalities or counties), which set different property tax rates for the same type of land use. Jurisdictions in the hinterlands often set lower rates than jurisdictions at the core of the urban area to attract more residents and firms. However, lower tax rates in the suburban fringe are very likely to act as levers for different manifestations of urban sprawl, especially decentralisation and low density. Empirical evidence from the US and Italy provides support to this argument (Ermini and Santolini, 2017; Song and Zenou, 2009). It is, thus, worth considering removing such incentives for decentralised and low-density development.

Consider using split-rate property taxes

A more targeted fiscal instrument to promote denser development is a split-rate, or two-rate, property tax, whereby higher tax rates are set on the value of land than on the value of buildings and other property improvements. Higher rates on land values in urban centres would discourage keeping land undeveloped (or underdeveloped) therein and, thus, reduce pressures on development at the rural-urban fringe.5 Such higher rates would also incentivise the redevelopment of urban brownfields, i.e. unused or underutilised commercial, industrial or residential sites where redevelopment is obstructed by contamination (see e.g. McCarthy, 2002; United States Environmental Protection Agency, 2017). Lower relative tax rates on the value of buildings and other improvements would further incentivise owners to build more intensively or renovate their properties to increase their value. That would enhance the base for property taxation in urban centres and reduce pressures on local finance (Gihring, 1999).

The location where split-rate taxes are implemented is a key determinant of their contribution to the control of urban sprawl (Banzhaf and Lavery, 2010; Gihring, 1999). For example, a combination of a (relatively) high tax on developed land and a (relatively) low tax on buildings at the urban fringe may incentivise the construction of higher buildings and be effective in decelerating the expansion of urban footprint. Such a split-rate tax could increase local density at the fringe, reducing this way the share of population residing in low-density areas and the portion of developed land hosting low density levels. On the other hand, the effect of a split-rate tax on fragmentation is unknown, as a substantial divergence of the two taxes may slow down any local infill development between existing fragments of urban fabric.

Despite their potential effectiveness in increasing urban density, the design and implementation of split-rate taxes is a challenging exercise. This is why they have seen very few applications thus far, and those are almost entirely concentrated in the United States. A major challenge in their application in policy practice is the separation of the value of land from the value of the improvements constructed on it. The implementation of split-rate taxes may also be deterred for political economy reasons (Cohen and Coughlin, 2005).

Develop incentive-based mechanisms to prevent the conversion of farmland and forestland

In many cities, the assessment of property values or the property tax rate varies according to land use. For example, preferential, or use-value, tax assessment is a commonly used instrument in the United States (Anderson, 1993). It provides tax incentives to owners of farmland, forestland and other types of undeveloped land to keep it in its current use, rather than sell it for urban development. This is achieved by taxing land at a lower value when it is used for e.g. agriculture or forestry than when it is allocated to residential or commercial uses (Bengston et al., 2004).

In other urban areas, the preservation of farmland and forestland is encouraged through lower tax rates on land used for agricultural or forestry purposes (Brandt, 2014). This instrument entails a direct link between the value of the property and the tax benefit that can be obtained from the preservation in its current use. Owners of properties of higher values, such as those close to urban centres or those near environmentally sensitive areas, will have stronger incentives to keep land in its current use. While this might be desirable for the latter, it may not be desirable for properties close to urban cores.

Similarly to the market-based instruments discussed above, spatial targeting is key for the success of policies aiming to prevent the conversion of farmland and forestland. Limiting the use of such incentive mechanisms to relatively undeveloped areas at the outskirts of the city, or close to environmentally sensitive areas (providing that current uses are sustainable and contribute to – rather than harm – the surrounding ecosystem) can prevent further fragmentation of urban development and decentralisation.

Consider developing policies based on development rights

Policies based on development rights are grounded in the idea that land ownership is equivalent to the ownership of a bundle of separable rights, such as mineral rights and development rights (Mills, 1980). Purchase of development rights is an instrument frequently used for the protection of open space. The land-owner sells the development rights to the government (or alternatively to a private non-profit organisation) and the land is permanently set aside for conservation purposes. From the government’s perspective, purchase of development rights can be much less expensive than public acquisition of land in areas where development pressures are low (Bengston et al., 2004).

A more relevant instrument to promote densification and reduce the fragmentation of urban fabric is, however, the transfer of development rights. Transferable, or tradable, development rights (TDRs) allow the transfer of density between properties. Almost all TDR schemes developed thus far have been set up in areas where some form of zoning policy had already been in place. More stringent development restrictions apply to the property selling the development rights, whereas the property purchasing the rights is allowed to be more densely developed than what is permitted under baseline zoning.6 Sending and receiving properties are usually located in different areas, but there are also cases where they are located in the same area (McConnell and Walls, 2009).7

TDR programmes are advocated for their potential to achieve the desired level of development – and therefore of open space conservation – in a cost-effective manner, and for their distributional benefits. For TDR schemes to have these desirable properties, key is the establishment of the correct cap, i.e. the total amount of development allowed by the scheme, and the allocation of a sufficient number of development rights to achieve that level of development.8 Trade of development rights then ensures that each land parcel is devoted to its most efficient use, provided that the market is sufficiently large and competitive (i.e. no participant has market power), and market participants have complete information (Levinson, 1997; McConnell and Walls, 2009; Mills, 1980). Benefits of TDR schemes are not limited to economic efficiency gains. In comparison with policies based on land-use regulation only (e.g. zoning), TDRs can also help address equity concerns. This is because they can lend themselves to compensation mechanisms for landowners whose properties are located in sending areas – areas where development is constrained for conservation or other purposes (Mills, 1980). What is more, compensation is not provided by scarce public funds, but instead through market forces. This attribute of development rights is clearly manifested in cases where they were provided in response to the re-zoning of specific areas. For example, they have been used to compensate landowners in areas which have been downzoned (McConnell and Walls, 2009).

TDR schemes can be particularly useful as a complement to other policies promoting densification, such as upzoning or minimum density restrictions. Nevertheless, the success of TDR programmes heavily relies on the level of activity in the TDR market. Market activity is generally higher when: (a) baseline zoning in sending areas is relatively stringent; and (b) demand for additional density in receiving areas – determined by both the discrepancy between desired and current density and the willingness of existing residents to accept higher density developments in their vicinity – is high.9 On the other hand, activity in the TDR market can be hampered by: (a) high transaction costs (e.g. pecuniary and time costs incurred in search of information about TDR prices);10 (b) additional requirements imposed on developers; and (c) the presence of other instruments allowing increased density in receiving areas (McConnell and Walls, 2009; Pruetz and Standridge, 2008).

Incentivise developers to provide public infrastructure for new constructions

More contiguous and centralised development can also be promoted through instruments requiring developers to cover – at least to some extent – the costs of providing public infrastructure to new constructions. Development impact fees are fees levied on developers to help pay for the capital costs of providing public infrastructure to new developments (Bengston et al., 2004; Juergensmeyer and Roberts, 2013, p. 319). Even though an instrument primarily used to finance the development of public infrastructure, development impact fees can act as powerful levers to direct urban expansion to areas closer to existing infrastructure (Bengston et al., 2004). As costs of constructing new infrastructure are lower in these areas, developers are indirectly incentivised to build in contiguous land.

Mitigating the environmental and economic consequences of sprawl-induced car use

Sustainable urban development cannot be achieved without greening urban transport systems and shifting travel demand towards other modes, such as public and non-motorised transport. Several dimensions of urban sprawl, such as low density, fragmentation and decentralisation, are strongly interlinked with car use, as it has been highlighted in Chapter 4. Travel distances are longer and the provision of public transport is significantly more costly in sprawling cities. This way, car becomes the most attractive, and for specific trip purposes, the only sensible transport alternative. Car dependency translates to immediate external effects, such as congestion and environmental externalities, which have to be mitigated. It also has long-run implications, as households and firms adjust their long-term decisions accordingly. That is, households will choose residences far from their workplace and shopping malls will be located far from their clients. This pattern creates a vicious circle where car dependency leads to even higher car use and additional needs for extensions of the road network and expansions of its capacity. This has important consequences for the environment and economic growth, manifested in significant increases in emissions, losses of open space and productive and valuable time lost in congestion.

The policies discussed here can lead to more sustainable urban transport patterns and mitigate some of the environmental and economic consequences of urban sprawl. Through changing the paradigm of urban mobility, they may also have significant long-term effects on the evolution of urban form itself. The discussion that follows focuses on a range of policies, including road pricing, parking policies, subsidies for public transport and motor fuel taxes. It also touches upon the potential effects of infrastructure investments, such as investments in road capacity expansion, on urban sprawl.

Introduce road pricing mechanisms

Motorists are rarely charged for the externalities they cause: congestion that implies time losses incurred by other road users; air pollution, with health consequences for the urban population; emissions of greenhouse gases, with global implications for climate. Congestion leads to notable losses of productive time and, therefore, to significant economic costs. Schrank et al. (2012) show that congestion caused 5.5 billion hours of delay in the United States in 2011, a number that roughly corresponds to a total time cost close to 0.9% of the GDP. In the same year, congestion caused a fuel waste of USD 121 billion, i.e. an additional 0.78% of the country’s GDP (not taking into account the associated external costs of fuel combustion). Regarding air pollution, it is well known that a substantial part of the emissions in urban areas originates from the use of private vehicles. The economic consequences of air pollution have been examined at the global scale in OECD (2016).

Efficient road pricing requires that the above costs are well reflected in road charges. As the aforementioned externality costs vary across road segments, charges should be higher in more congested and more densely populated parts of urban space, and vice versa. In addition, an ideal pricing scheme should also be time-variant, as traffic levels vary across time intervals of a given day and across days of the week. This ultimately entails that congestion charges should vary across routes and time intervals (Anas and Lindsey, 2011). As air emissions largely depend on fuel type, pollution control technologies, the weight and vintage of the car, optimal charges should also vary by car model and vintage.

Apart from mitigating the direct consequences of unsustainable urban patterns, road pricing mechanisms may provide long-run incentives for denser, more contiguous development, as they could shorten distances between residential locations and key points of economic activity. On top of the spatial configuration of a given area, the long-run effect of road pricing on urban form depends on two important parameters. First, the technical features of the implemented scheme determine who is going to be affected by it. That is, the extent to which the scheme obliges drivers to internalise the external effects of car use determines the size of the population groups for which short-run responses to it (e.g. switching to public transport, shorter shopping and leisure trips with car) and long-run adjustments (e.g. household relocation, change of employment location) are relevant. A pricing scheme is therefore more likely to affect urban form (in the long run) when it affects the behaviour of a relatively larger population group, i.e. when it leaves a relatively smaller portion of traffic unpriced. Second, the effect of a road pricing policy on urban form depends on its potential to induce long-run adjustments to the behaviour of the affected groups. In that sense, permanent road pricing schemes may induce higher density levels around major employment locations, especially if the regulatory mechanisms in land and housing markets are laxer.11

Optimal road pricing instruments are very effective in mitigating traffic externalities and in greening urban transport. In addition, they may play an auxiliary role in inducing desirable changes in urban form in the long run. However, time- and space-variant optimal road charges are costly to implement in practice, mainly due to the large costs of installing the required infrastructure. Therefore, road pricing schemes that are simpler in administrative terms may currently be more feasible to implement. The following analysis focuses on two such types of second-best road pricing schemes, i.e. cordon tolls and flat kilometre taxes, their potential to mitigate the external effects of car use (mainly air pollution and congestion) and the extent to which they could influence the evolution of urban form, as the latter was presented in Chapter 2 and measured in Chapter 3.

Cordon tolls

Cordon tolls charge all inbound (and sometimes also outbound) trips to a certain bounded area, which usually encompasses a central business district. These charges are distance-invariant, in the sense that a given driver pays the same fee when entering or exiting the bounded area, regardless of the total length of the trip. In general, cordon tolls are more effective in reducing congestion in urban settings resembling monocentric cities, in which the vast majority of peak-hour car trips are destined to locations inside the cordon (Mun et al., 2003, 2005; Verhoef, 2005). The measure has been shown to reduce car traffic considerably. For example, a fee on cars entering or exiting Stockholm’s inner city has been shown to reduce car traffic around charging points by 22% (Eliasson et al., 2009). Because congestion is a non-linear phenomenon, i.e. small reductions in traffic levels may give rise to substantial time savings, travel times in parts of the network fell by more than 22%. From an environmental point of view, the cordon toll reduces aggregate emissions by inducing changes in the modal split, i.e. by inducing some car users to switch to public transport or other modes. It also increases urban travel speeds to levels where fuel consumption per kilometre is lower. Because in many cases the time savings are associated with fewer traffic jams occurring in network bottlenecks, the scheme may help reduce the non-exhaust emissions arising from queuing (e.g. wear and tear in brakes and tyres, which is more intense with start-and-stop driving). In Stockholm, the reduction in air pollutant concentrations inside the cordon was estimated to lie between 10% and 14% (depending on the pollutant), although it was found to be substantially smaller for nitrogen oxides (8.5%).

Policy makers should also keep in mind that a cordon toll increases the accessibility of locations lying within the bounded area, as all trips with origin and destination within that area go uncharged. Therefore, the measure creates a long run centripetal force which increases demand for land and housing (and, in turn, their respective prices) within the cordon. Therefore, the ceteris paribus effect of that force on average population density could be significantly positive, provided that housing supply within the cordon is elastic. The effect on the allocation indicators presented in Chapter 2 is more perplexed. That is, cordon tolls are expected to increase the average population density in locations within the cordon and reduce it in locations outside of it, but the overall effect on the two indicators depends on the relative change between the two areas. Finally, with such a policy all traffic in locations outside the cordoned area goes unpriced. Thus the policy per se should not be expected to affect fragmentation levels in periurban areas significantly.

Flat kilometre taxes

The drawbacks of a cordon toll scheme call for a distance-based charge that accounts for the strong positive correlation between the external costs of a trip and its length. The recent technological advances made the use of Global Positioning Systems (GPS) with tracking devices cheaper and easier to implement. These technologies have made the introduction of a flat kilometre tax in an urban area a meaningful alternative to a cordon toll. Under this scheme, the use of private vehicles is charged on a distance basis, but only within the predetermined boundaries of the implementation zone. The tax is flat in the sense that it does not vary across space or over time. From an environmental point of view, the measure complements motor fuel taxes: it increases the kilometre costs of car use in areas where the external costs of it (congestion, pollution) are particularly high and allows motor fuel taxes to be used for pricing carbon emissions and possibly other policy purposes.

The long-run effect of a flat kilometre tax on urban form is very different from that of a cordon toll. Ceteris paribus, the former scheme contributes to more compact development, in which jobs, residences and key points of economic activity lie closer to each other. In monocentric settings, this translates to an overall densification (higher average population density) and a decrease in the share of population residing in areas of very low density (below 1 500 inhabitants per km2) Furthermore, distance-based charging in monocentric settings is expected to decrease fragmentation, because leapfrog development, which is usually occurring in remote areas, becomes more expensive. In polycentric cities, distance-based charges may have significantly different effects.

Possible distributional effects of road pricing schemes, as well as interactions between road pricing and the fiscal system should be taken into consideration prior to implementation. Specific types of road pricing, such as cordon tolls, are likely to create winners and losers or distribute their welfare gains unequally. In these cases, recycling revenue in a way that favours the affected areas and population groups, such as earmarking revenues to subsidise public transport, may be warranted. Furthermore, labour supply from areas with limited accessibility to public transport is likely to be affected, especially if the toll levels are set high enough to reflect the social costs of generalised private vehicle use in these areas. The erosion of the labour income tax base due to a road toll in the commuting hours may at first appear as small. However, numerical simulations show that in a highly-distortionary fiscal framework it may be large enough to neutralise the welfare gains from the mitigation of the traffic-related externalities (Parry and Bento, 2001; Tikoudis, Verhoef and van Ommeren, 2015). To circumvent this issue, the aforementioned groups should be identified prior to the implementation of the scheme and offered a lower toll during the peak commuting hours.

In summary, policy makers should consider the introduction of road pricing as a meaningful alternative to massive investments in roads and highways, which have characterised public policy during the last decades. Urban congestion management is a meaningful alternative which, as demonstrated by the steadily increasing number of cities adopting it (e.g. London, Milan, Oslo, Singapore, Stockholm), is gaining momentum around the world. Road pricing, i.e. charges reflecting the costs of congestion, air pollution and other externalities of private vehicle use, is a very powerful policy instrument to influence travel behaviour and location choices. In the long run, road pricing is likely to incentivise infill development, encourage more polycentric structures and promote densification of urban areas. Those effects will be significantly stronger when road pricing is combined with efficient motor fuel taxes.

Increase motor fuel taxes to account for the marginal external costs of fuel consumption

Motor fuel consumption has significant environmental costs, caused by emissions of greenhouse gases and air pollutants. Despite the widespread use of motor fuel taxes in OECD countries and beyond, they are in many cases set at relatively low levels, which do not fully cover the marginal external costs of fuel consumption.12 This is especially the case in countries where road pricing does not exist and motor fuel taxes are supposed to also reflect other externalities of car use (e.g. congestion). In most countries, taxes on diesel fuel are lower than those on gasoline, inducing more travel in (usually more polluting) diesel cars (Harding, 2014). Setting motor fuel taxes at lower than desirable levels has similar effects on car use and urban form as leaving congestion unpriced. It leads to excessive car use and in the long-run to more sprawling city structures, characterised by sparser and more decentralised development. It is thus important that motor fuel taxes are set at levels fully accounting for the environmental costs of fuel consumption and, in combination with road pricing, incentivise more sustainable urban development patterns.

Reform parking policy

Urban areas allocate enormous amounts of land to car parking. Parking spaces are provided on street, in residential and commercial buildings, in car parks and elsewhere. High parking supply not only induces expansion of the urban fabric because it consumes land that could be allocated to other – potentially more productive – uses, but also because it encourages car ownership and use. Households with easy access to parking spaces close to their residence and workplace are likely to own more cars than households who do not (De Groote, van Ommeren and Koster, 2016; Guo, 2013). In addition, the availability of a parking space at the origin and/or destination is an important determinant of one’s decision to make a trip by car rather than by another transport mode (Weinberger, 2012).

Parking policy does not only influence households’ modal choice: it also affects their choice of residence location. Widespread availability of low-priced parking spaces may well induce households to move further away from business centres as they can conveniently rely on their cars to cover their everyday needs.

Consider removing minimum parking requirements and replacing them with maximum ones

Policies aimed at regulating the supply of parking spaces have important effects on how an urban area is shaped and how car-dependent it becomes. Minimum parking requirements for residential and commercial buildings are widespread in OECD countries and beyond. In practice, this regulatory instrument usually requires developers to provide a sufficient number of parking spaces to cover peak demand for free parking (Shoup, 1999a). In most cases, minimum parking requirements were established to provide low-priced off‐street parking to motorists, so that local congestion (caused by cars cruising to find a parking space) is prevented and local business is stimulated (Shoup, 1999a, 1999b). However, minimum parking requirements result in more parking spaces per parcel than what would have been provided if developers were free to decide how many spaces they would offer (Cutter and Franco, 2012). They drive building costs up and, therefore, discourage new development (Shoup, 1999a). They also encourage car use, as the oversupply of parking spaces decreases the total costs of urban trips by car (Manville, 2013). In sum, minimum parking requirements are very likely to lead eventually to less dense and more decentralised cities and unsustainable urban development (Willson, 1995). Therefore, it is worth considering the removal of minimum parking requirements for new developments.

The unintended consequences of minimum off-street parking requirements have induced policy makers to seek alternative instruments to regulate parking. Several major cities in the OECD, including London, New York City and Paris, have applied some form of maximum parking requirements for particular land uses. Evidence from London’s major parking policy reform in 2004 shows that the replacement of minimum parking requirements with maximum ones led to a notable 49% reduction of parking spaces in new developments. The largest part of that change was due to the removal of the minimum parking requirements, which particularly affected developments in Inner London. In contrast, maximum parking requirements mainly affected new developments in London’s suburbs (Li and Guo, 2014).

Price on-street parking appropriately

Parking policy is mainly based on the use of regulatory instruments. If on-street parking were priced at its marginal social cost, i.e. the sum of the costs of providing a parking space and the external costs of that space, the need for regulatory policies on off‐street parking would be almost completely eliminated. The external costs of parking provision comprise the external costs of cruising for parking, stemming from the time losses caused by searching for a space, and the external costs of open space lost to convert land to parking space (see also Inci, 2015). External costs of congestion and air emissions vary across space and over time and, hence, parking tariffs should ideally be space- and time-variant (Vickrey, 1954, cited in Inci, 2015). Perhaps the most well-known real-world application of space- and time-varying parking tariffs is San Francisco’s SFpark programme (see Pierce and Shoup, 2013).

In reality, however, on-street parking is significantly underpriced, and prices of parking spaces rarely cover the costs of provision – let alone external costs. The underpricing of on‐street parking is well manifested in the important price differences existing in most cities between on-street and garage parking. If prices were equal, cruising for parking would, at least in theory, be eliminated (Inci, 2015). The underpricing of on-street parking can have important implications for urban sprawl, as it reduces the monetary costs of car trips, and therefore encourages car dependency.13 Car dependency encourages lower density and more decentralised development, as households are anyway used to cover virtually all their travel needs by car. On-street parking prices reflecting the full costs of parking provision – including external costs – is an important instrument in policy makers’ hands to control urban sprawl, raise public revenues and promote sustainable urban development.

Shift investments to more sustainable forms of transport infrastructure

Excessive investment in road transport infrastructure is responsible for more car use, higher emissions from road transport and urban sprawl. For example, it has been shown that the construction of highways induces urban areas to sprawl to remote locations and become more fragmented and decentralised (Garcia-López, Holl, and Viladecans-Marsal, 2015; Garcia-López, Solé-Ollé, and Viladecans-Marsal, 2015; Su and DeSalvo, 2008).

Shifting investments from new road transport infrastructure to public transport or to infrastructure for non-motorised modes (e.g. bikeways and sidewalks) is likely to lead in the long run to substantially less car use, deter further urban sprawl and induce infill development. It may also lead to more polycentric urban structures. The effects of this shift of investments will be reinforced when it is accompanied by efficient pricing of congestion, on-street parking and motor fuel consumption.

Increasing effectiveness and public acceptance of policies to mitigate urban sprawl’s consequences

Regardless of the policy instrument used to control urban sprawl or mitigate its consequences, a number of actions can help design better policies and increase their public acceptance. For example, policy makers should take care to design policies which can control sprawl without impeding urbanisation, consider the long-term effects of proposed policies, ensure that they have well understood their possible distributional consequences and how they will be addressed, and be aware of their fiscal implications.

Control urban sprawl without hampering urbanisation

To address concerns over urban sprawl, it is important to understand the differences between that phenomenon and the phenomenon of urbanisation and identify whether urban development patterns are following an unsustainable path in a particular city context. This report can contribute to such efforts, as it indicates urban areas where development patterns may need increased attention from policy makers (see Chapter 3). In general, policies should focus on mitigating the consequences of urban sprawl without hampering urbanisation. Policies discouraging urbanisation may be detrimental for economic growth, as the latter has been shown to be fostered by the positive externalities arising from the clustering of people and firms in space, i.e. from the formation and growth of urban areas. By treating urban sprawl as a problem of city growth, policy makers may end up favouring interventions, such as restrictions in housing supply, that attempt to curtail the latter without offering a remedy to the environmental and economic consequences of urban sprawl.

Consider carefully the future social costs and benefits of land-use policy decisions

Policies implemented today will determine social costs and benefits in the medium and long run. Spatial planning and land-use policies have persistent long-term effects: land-use decisions taken today will lock cities in specific development patterns for many years to come. For example, postponing densification policies implies a higher future social cost of public transport provision and a lower frequency of service in a larger part of the urban area. In turn, that increases the likelihood of future adoption of policies favouring the use of private vehicles, such as investments in new roads. It is, thus, of primary importance that the long-term social benefits and costs of alternative urban policies are well understood before relevant decisions are made.

Share positive and negative experiences with policies to control urban sprawl

The environmental, economic and social effects of urban policies are highly context-specific. Although the same principles may apply to different urban settings, policies which are proven to be effective in one urban context will not necessarily be effective in other contexts. However, there are important lessons to be learnt from other cities’ experiences with specific urban policies. A city may be similar to others in a number of aspects and, thus, policies which work in that city could also work in cities which are relatively similar to it. It is important that city authorities engage in further sharing of their urban policy experiences and highlight the policies that have been effective in achieving their objectives, as well as the ones that failed to do so.

Engage the public and relevant stakeholders

Communicating clearly the long-term benefits of green urban policies to the public and relevant stakeholders can increase support for these policies. In most cases, the public and other stakeholders may not be aware of the various environmental and economic consequences of rapid, uncontrolled expansion of the urban fabric. Furthermore, the vast majority of the population is not aware of the fiscal challenges related to providing low-density areas with public services. Therefore, public support of many of the seemingly unpopular policies aiming to control urban sprawl could be increased once their net social benefits are clearly conveyed to the public and other stakeholders. At the same time, consultation of the public and relevant stakeholders at an early stage of decision-making ensures that their voices and concerns are heard and facilitates the identification of possible complementary measures needed to compensate the losers from the policy intervention.14

Ensure that potential distributional consequences of green urban policies are minimised

Policies to control urban sprawl or mitigate its negative effects are sometimes abandoned due to concerns over their possible distributional consequences. However, these concerns are not always well-founded and in many cases distributional effects are overplayed. More importantly, even when green urban policies are anticipated to place a disproportionate burden on more vulnerable population groups, this burden can be removed by targeted compensatory mechanisms. For example, part of the revenues raised from tax-based instruments could be used to provide direct financial support to vulnerable households, e.g. in the form of tax credits. Another mechanism could involve earmarking a part of the revenues to develop policies that promote environmental quality in the areas where more vulnerable groups reside or to finance improvements in the function of the public transport system.

Consider whether policy instruments can contribute to fiscal consolidation

Green urban policies can also help governments at all levels consolidate their budgets. Unsustainable urban development patterns are often responsible for significant pressures on public finance. For example, providing public services to low-density or remote areas can be especially challenging from a financial perspective, and can even lead to large deficits. In many cities, the social cost of providing services in such areas should be internalised in development costs and housing prices to a greater extent. National and local governments should reconsider the substantial subsidies required to maintain the provision of public transport, water, sewerage and other goods and services, and seek alternative financing mechanisms that reflect better the external costs of urban sprawl (e.g. development impact fees).

Be committed to implemented policies

For urban policies to be successful, it is necessary that they are not perceived as short-term remedies which will be discontinued in the longer term. Governments should show commitment to implemented policies, so that all stakeholders are convinced to take them into account when making longer-term decisions. Given the time required by most urban land-use and transport policies for their long-run effects to kick in, it is important that government support of them extends beyond one or more terms of office.

5.3. Following an integrated approach to make urban development more sustainable

Urban development is heavily influenced by policy interactions, both between sectoral policies and between policies implemented by different jurisdictions. Taking these interactions into account before policy action is a key determinant of arriving at the right policy mix for sustainable urban development. This section discusses two types of policy interactions affecting urban development: i) interactions between policies developed by different jurisdictions; and ii) interactions between sectoral policies, e.g. between land-use and transport policies. It then highlights the need for an integrated approach to policy-making to control urban sprawl and mitigate its consequences.

Interactions between policies developed by different jurisdictions

The consequences of urban sprawl would ideally be tackled with policies at the urban area level. However, this is rarely the case in practice for two reasons. First, urban areas are seldom defined within the boundaries of a single jurisdiction. They usually encompass multiple jurisdictions which have the authority to develop and implement policies within their own administrative boundaries. Despite their good intentions, unilateral policies to control urban sprawl by a certain jurisdiction are likely to be ineffective as population growth and development may shift to neighbouring jurisdictions in the same urban area. Second, urban areas are not static structures; they are dynamic organisms whose boundaries evolve over time. The boundaries of administrative powers, on the other hand, do not usually change in a timely manner with the evolution of urban development.

Decentralised governance systems – where local governments have the power to set land-use policies within their own jurisdiction – have been shown to provide implicit incentives for urban sprawl. In contrast, more centralised governance systems – where land-use policy is set at the national, regional or state level – are more likely to lead to more compact urban development, but also to insufficient supply of housing (Blöchliger et al., 2017).

Two main reasons can explain the influence of decentralised systems on urban sprawl. First, developers can circumvent stringent local land-use policies by shifting to nearby jurisdictions with laxer policy frameworks. This induces leapfrog development, which can result in a more sprawled urban area (Blöchliger et al., 2017; Glaeser and Kahn, 2004). Second, in more decentralised settings, homeowners may have the power to influence local land-use policies for their benefit. For example, homeowners have incentives to lobby for more stringent zoning rules which will attract high-income households in their neighbourhood and increase the prices of their dwellings (a process which is also known as fiscal zoning). Such ordinances are likely to induce low-density urban development and social segregation (Blöchliger et al., 2017; Hilber and Robert-Nicoud, 2013).

In general, the alignment of policies implemented by neighbouring local authorities can be key for controlling urban sprawl and tackling its consequences. Each urban area has a unique spatial and institutional configuration, but it is important that governments at higher levels, such as national and regional/state governments, have a complete understanding of the environmental and economic implications of that configuration. In some urban areas, national and/or regional authorities have endorsed the establishment of formal clusters of neighbouring local governments, who have the overall responsibility for spatial planning and the provision of public services in the urban area. This could also be something worth considering in the case of other cities. Under specific circumstances, it may also be meaningful for higher-tier governments to consider the compilation of national or regional urban policy frameworks with clear environmental and economic objectives. Regardless of the governance system, guidance from national or regional authorities can significantly contribute to the development of more sustainable urban policies.

At the same time, it is important that national, regional and local policies to control urban sprawl and mitigate its effects are better aligned. Several policy instruments controlled by national or regional governments often trigger similar behavioural responses to those induced by instruments controlled by local authorities. Even worse, the efficacy of policies developed by an authority may be undermined by policies implemented by authorities at other levels of government. Policies designed by different government levels usually serve very diverse objectives, but their design could be reconsidered to discourage undesirable urban development patterns.

Interactions between sectoral policies

The incentives provided by land-use, transport and environmental policies may reinforce or offset each other. Effective long-term urban planning implies that interactions between sectoral policies are clearly understood and taken into account, but sectoral policy often considers instruments in isolation from the others. For example, the role of massive investments in urban highways in generating incentives for low-density development has, in many cases, been persistently neglected by policy makers in OECD countries. Similarly, urban planning policies have neglected the pressures they may impose on the transport system. These pressures should be reflected in appropriate urban planning policy instruments.

An integrated approach towards more sustainable urban development

It is important that policy makers undertake the challenging task of developing integrated solutions to control the various manifestations of urban sprawl and mitigate its effects. An integrated approach should be followed at all stages of policy-making: identification of the problem that needs to be tackled by policy intervention, policy design, implementation and evaluation. This integrated approach is at the heart of concepts which have often been used as paradigms for future urban development, such as the compact city or the smart city. A relatively more concrete integrated approach that cities could be inspired from is transit-oriented development.

Compact, high-density, mixed-use development around major transport nodes may be an answer to multiple economic and environmental challenges emerging in urban areas. Transit-oriented development, hereafter TOD, can significantly reduce car dependency, make the public transport system and other key points of interest accessible to everyone, sharply decrease the costs of the provision of public services, protect natural amenities and periurban ecosystems, increase proximity to open spaces, reduce population exposure to air pollution and mitigate the problem of affordable housing.

Instruments that can be used to promote TOD, and are critical for its effectiveness in achieving the objectives above, include minimum height restrictions in undeveloped periurban areas to be conceded for future development, appropriate pricing of car use (including pricing for congestion, parking and the environmental costs of car use), measures to prioritise densification around key nodes of the existing public transport network, and a shift of investments from roads to urban public transport infrastructure.

5.4. Conclusions

Sprawling urban development patterns have multiple private benefits, which should, however, be weighed against their social costs. Market forces fail to take these social costs into account and, thus, policy intervention is in many cases necessary to direct urban development to more sustainable patterns. Acknowledging that developing policies to this end is very challenging and highly context-specific, the chapter discussed a number of land-use, transport and fiscal policy reforms that can be relevant in certain urban settings.

The chapter also highlighted the need for an integrated policy approach to understanding the issues at stake and developing solutions to address them. This integrated approach would also involve identifying and taking into account relevant policy interactions, both between sectoral policies and between policies implemented by different jurisdictions, before making relevant decisions. In any case, policy development should be based on a mix of instruments which would provide both immediate remedies to the consequences of harmful development patterns and long-term solutions to unsustainable urban expansion. Policy instruments targeted to greening urban transport can be particularly effective in the short run, while land-use policy reforms can pay off in the longer run.


Amati, M. (2016), “Green Belts: A Twentieth-century Planning Experiment”, in Amati M. (Ed.), Urban Green Belts in the Twenty-first Century, Routledge, Abingdon, UK.

Amati, M. and M. Yokohari (2006), “Temporal changes and local variations in the functions of London’s green belt”, Landscape and Urban Planning, Vol. 75/1, pp. 125-142.

Anas, A. and R. Lindsey (2011), “Reducing Urban Road Transportation Externalities: Road Pricing in Theory and in Practice”, Review of Environmental Economics and Policy, Vol. 5/1, pp. 66-88.

Anderson, J.E. (1993), “Use-Value Property Tax Assessment: Effects on Land Development”, Land Economics, Vol. 69/3, pp. 263-269.

Bae, C.-H.C. and M.-J. Jun (2003), “Counterfactual Planning”, Journal of Planning Education and Research, Vol. 22/4, pp. 374-383.

Ball, M. et al. (2014), “Urban growth boundaries and their impact on land prices”, Environment and Planning A, Vol. 46/12, pp. 3010-3026.

Banzhaf, H.S. and N. Lavery (2010), “Can the land tax help curb urban sprawl? Evidence from growth patterns in Pennsylvania”, Journal of Urban Economics, Vol. 67/2, 169-179.

Bengston, D.N., J.O. Fletcher and K.C. Nelson (2004), “Public policies for managing urban growth and protecting open space: policy instruments and lessons learned in the United States”, Landscape and Urban Planning, Vol. 69/2, pp. 271-286.

Bengston, D.N. and Y.-C. Youn (2006), “Urban containment policies and the protection of natural areas: The case of Seoul’s greenbelt”, Ecology and Society, Vol. 11/1, pp. 3.

Bertaud, A. and J.K. Brueckner (2005), “Analyzing building-height restrictions: Predicted impacts and welfare costs”, Regional Science and Urban Economics, Vol. 35/2, pp. 109-125.

Bird, R.M. and E. Slack (2004), “Land and property taxation in 25 countries: A comparative review”, in Bird, R.M. and E. Slack (eds), International Handbook of Land and Property Taxation, Edward Elgar, Cheltenham, UK, and Northampton, MA, USA.

Blöchliger, H., et al. (2017), “Local taxation, land use regulation, and land use: A survey of the evidence”, OECD Economics Department Working Papers, No. 1375, OECD Publishing, Paris,

Brandt, N. (2014), “Greening the Property Tax”, OECD Working Papers on Fiscal Federalism, No. 17, OECD Publishing, Paris,

Brueckner, J.K. and H.-A. Kim (2003), “Urban Sprawl and the Property Tax”, International Tax and Public Finance, Vol. 10/1, pp. 5-23.

Brueckner, J.K. and K.S. Sridhar (2012), “Measuring welfare gains from relaxation of land-use restrictions: The case of India’s building-height limits”, Regional Science and Urban Economics, Vol. 42/6, pp. 1061-1067.

Caldera Sánchez, A. and Å. Johansson (2011), “The Price Responsiveness of Housing Supply in OECD Countries”, OECD Economics Department Working Papers, No. 837, OECD Publishing, Paris,

Cho, C.-J. (2002), “The Korean growth-management programs: Issues, problems and possible reforms”, Land Use Policy, Vol. 19/1, pp. 13-27.

Cohen, J.P. and C.C. Coughlin (2005), “An Introduction to Two-Rate Taxation of Land and Buildings”, Federal Reserve Bank of St. Louis Review, Vol. 87/3, pp. 359-374.

Cutter, W.B. and S.F. Franco (2012), “Do parking requirements significantly increase the area dedicated to parking? A test of the effect of parking requirements values in Los Angeles County”, Transportation Research Part A: Policy and Practice, Vol. 46/6, pp. 901-925.

De Groote, J., J.N. van Ommeren and H.R.A. Koster, (2016), “Car ownership and residential parking subsidies: Evidence from Amsterdam”, Economics of Transportation, Vol. 6, pp. 25-37.

Eliasson, J. et al. (2009), “The Stockholm congestion-charging trial 2006: Overview of effects”, Transportation Research Part A: Policy and Practice, Vol. 43/3, pp. 240-250.

Ermini, B. and R. Santolini (2016), “Urban sprawl and property tax of a city’s core and suburbs: Evidence from Italy”, Regional Studies, Vol. 51/9, pp. 1374-1386.

Fischel, W.A. (2001), “Why are there NIMBYs?”, Land Economics, Vol. 77/1, pp. 144-152.

Garcia-López, M.-Á., A. Holl and E. Viladecans-Marsal (2015), “Suburbanization and highways in Spain when the Romans and the Bourbons still shape its cities”, Journal of Urban Economics, Vol. 85, pp. 52-67.

Garcia-López, M.-Á., A. Solé-Ollé and E. Viladecans-Marsal (2015), “Does zoning follow highways?”, Regional Science and Urban Economics, Vol. 53, pp. 148-155.

Gennaio, M.-P., A.M. Hersperger and M. Bürgi (2009), “Containing urban sprawl – Evaluating effectiveness of urban growth boundaries set by the Swiss Land Use Plan”, Land Use Policy, Vol. 26/2, pp. 224-232.

George, H. (1879), Progress and Poverty: An Inquiry into the Cause of Industrial Depressions and of Increase of Want with Increase of Wealth: The Remedy, Doubleday, Page & Co, Garden City, NY.

Gihring, T.A. (1999), “Incentive property taxation: A potential tool for urban growth management”, Journal of the American Planning Association, Vol. 65/1, pp. 62-79.

Glaeser, E.L. and M.E. Kahn (2004), “Sprawl and urban growth”, in Henderson, V. and J.-F. Thisse (Eds.), Handbook of Regional and Urban Economics, Vol. 4, Cities and Geography, Elsevier, Amsterdam, pp. 2481-2527.

Guo, Z. (2013), “Does residential parking supply affect household car ownership? The case of New York City”, Journal of Transport Geography, Vol. 26, pp. 18-28.

Harding, M. (2014), “The Diesel Differential: Differences in the Tax Treatment of Gasoline and Diesel for Road Use”, OECD Taxation Working Papers, No. 21, OECD Publishing, Paris,

Harman, B.P., R. Pruetz and P. Houston (2015), “Tradeable development rights to protect peri-urban areas: Lessons from the United States and observations on Australian practice”, Journal of Environmental Planning and Management, Vol. 58/2, pp. 357-381.

Hilber, C.A.L. and F. Robert-Nicoud (2013), “On the origins of land use regulations: Theory and evidence from US metro areas”, Journal of Urban Economics, Vol. 75, pp. 29-43.

Ihlanfeldt, K.R. (2007), “The effect of land use regulation on housing and land prices”, Journal of Urban Economics, Vol. 61/3, pp. 420-435.

Inci, E. (2015), “A review of the economics of parking”, Economics of Transportation, Vol. 4/1-2, pp. 50-63.

Janssen-Jansen, L.B. (2008), “Space for Space, a transferable development rights initiative for changing the Dutch landscape”, Landscape and Urban Planning, Vol. 87/3, pp. 192-200.

Juergensmeyer, J.C. and T. Roberts (2013), Land Use Planning and Development Regulation Law, Third Edition, Hornbook Series, West Academic, St. Paul, MN.

Jun, M.-J. (2004), “The effects of Portland’s urban growth boundary on urban development patterns and commuting”, Urban Studies, Vol. 41/7, pp. 1333-1348.

Lee, C.-M. (1999), “An intertemporal efficiency test of a greenbelt: Assessing the economic impacts of Seoul’s greenbelt”, Journal of Planning Education and Research, Vol. 19/1, pp. 41-52.

Lee, C.-M. and P. Linneman (1998), “Dynamics of the greenbelt amenity effect on the land market – The case of Seoul’s greenbelt”, Real Estate Economics, Vol. 26/1, pp. 107-129.

Levinson, A. (1997), “Why oppose TDRs? Transferable development rights can increase overall development”, Regional Science and Urban Economics, Vol. 27/3, pp. 283-296.

Li, F. and Z. Guo (2014), “Do parking standards matter? Evaluating the London parking reform with a matched-pair approach”, Transportation Research Part A: Policy and Practice, Vol. 67, pp. 352-365.

Manville, M. (2013), “Parking requirements and housing development: Regulation and reform in Los Angeles”, Journal of the American Planning Association, Vol. 79/1, pp. 49-66.

McCarthy, L. (2002), “The brownfield dual land-use policy challenge: Reducing barriers to private redevelopment while connecting reuse to broader community goals”, Land Use Policy, Vol. 19/4, pp. 287-296.

McConnell, V. and M. Walls (2009), “Policy Monitor: U.S. Experience with Transferable Development Rights”, Review of Environmental Economics and Policy, Vol. 3/2, pp. 288-303.

Micelli, E. (2002), “Development rights markets to manage urban plans in Italy”, Urban Studies, Vol. 39/1, pp. 141-154.

Mills, D.E. (1980), “Transferable development rights markets”, Journal of Urban Economics, Vol. 7/1, pp. 63-74.

Mirrlees et al. (2011), “The Taxation of Land and Property”, in Mirrlees et al. (ed), Tax by Design: The Mirrlees Review, Oxford University Press, Oxford.

Mun, S.I., K.J. Konishi and K. Yoshikawa (2005), “Optimal cordon pricing in a non-monocentric city”, Transportation Research Part A: Policy and Practice, Vol. 39/7, pp. 723-736.

Mun, S.I., K.J. Konishi and K. Yoshikawa (2003), “Optimal cordon pricing”, Journal of Urban Economics, Vol. 54/1, pp. 21-38.

Parry, I.W. and A. Bento (2001), “Revenue recycling and the welfare effects of road pricing”, The Scandinavian Journal of Economics, Vol. 103/4, pp. 645-671.

Pierce, G. and D.C. Shoup (2013), “Getting the prices right: An evaluation of pricing parking by demand in San Francisco”, Journal of the American Planning Association, Vol. 79/1, pp. 67-81.

Pruetz, R. and N. Standridge (2008), “What makes transfer of development rights work? Success factors from research and practice”, Journal of the American Planning Association, Vol. 75/1, pp. 78-87.

Schrank, D., B. Eisele and T. Lomax (2012), TTI’s 2012 Urban Mobility Report, Texas A&M Transportation Institute, The Texas A&M University System, College Station, TX.

Schuetz, J. (2009), “No renters in my suburban backyard: Land use regulation and rental housing”, Journal of Policy Analysis and Management, Vol. 28/2, pp. 296-320.

Shoup, D.C. (1999a), “The trouble with minimum parking requirements”, Transportation Research Part A: Policy and Practice, Vol. 33/7, pp. 549-574.

Shoup, D.C. (1999b), “In Lieu of Required Parking”, Journal of Planning Education and Research, Vol. 18/4, pp. 307-320.

Siedentop, S., S. Fina and A. Krehl (2016), “Greenbelts in Germany’s regional plans – An effective growth management policy?”, Landscape and Urban Planning, Vol. 145, pp. 71-82.

Silva, E. and R. Acheampong (2015), “Developing an Inventory and Typology of Land-Use Planning Systems and Policy Instruments in OECD Countries”, OECD Environment Working Papers, No. 94, OECD Publishing, Paris,

Small, K.A. and S. Song (1994), “Population and employment densities: Structure and change”, Journal of Urban Economics, Vol. 36/3, pp. 292-313.

Song, Y. and Y. Zenou (2009), “How do differences in property taxes within cities affect urban sprawl?”, Journal of Regional Science, Vol. 49/5, pp. 801-831.

Song, Y. and Y. Zenou (2006), “Property tax and urban sprawl: Theory and implications for US cities”, Journal of Urban Economics, Vol. 60/3, pp. 519-534.

Su, Q. and J.S. DeSalvo (2008), “The effect of transportation subsidies on urban sprawl”, Journal of Regional Science, Vol. 48/3, pp. 567-594.

Taylor, J., C. Paine and J. FitzGibbon (1995), “From greenbelt to greenways: Four Canadian case studies”, Landscape and Urban Planning, Vol. 33/1-3, pp. 47-64.

Tikoudis, I., E.T. Verhoef and J.N. van Ommeren (2015), “On revenue recycling and the welfare effects of second-best congestion pricing in a monocentric city”, Journal of Urban Economics, Vol. 89, pp. 32-47.

United Nations (2015), Transforming our World: The 2030 Agenda for Sustainable Development, Resolution adopted by the General Assembly on 25 September 2015. A/RES/70/1.

UNECE (1998), Convention on Access to Information, Public Participation in Decision-making and Access to Justice in Environmental Matters, Aarhus, Denmark, 25 June.

United States Environmental Protection Agency (2017), Brownfield Overview and Definition. Retrieved from: (accessed 3 March 2017),

Verhoef, E.T. (2005), “Second-best congestion pricing schemes in the monocentric city”, Journal of Urban Economics, Vol. 58/3, pp. 367-388.

Vickrey, W.S. (1996), “Simplification, Progression and a Level Playing Field”, Columbia University Discussion Paper Series, No. 9596-5, New York, NY.

Vickrey, W.S. (1954), “The economizing of curb parking space”, Traffic Engineering, Vol. 25/2, pp. 62-67.

Vyn, R.J. (2012), “Examining for Evidence of the Leapfrog Effect in the Context of Strict Agricultural Zoning”, Land Economics, Vol. 88/3, pp. 457-477.

Wassmer, R.W. (2016), “Further empirical evidence on residential property taxation and the occurrence of urban sprawl”, Regional Science and Urban Economics, Vol. 61, pp. 73-85.

Weinberger, R. (2012), “Death by a thousand curb-cuts: Evidence on the effect of minimum parking requirements on the choice to drive”, Transport Policy, Vol. 20, pp. 93-102.

Willson, R.W. (1995), “Suburban parking requirements: A tacit policy for automobile use and sprawl”, Journal of the American Planning Association, Vol. 61/1, pp. 29-42.


← 1. Periodical reviews of greenbelts may also be useful for another reason. As congestible local public goods, the total value of the amenity they provide increases with the total number of people enjoying it. However, many of the ecosystem and recreational services provided by greenbelts are prone to congestion, and accessibility to the greenbelt itself will decrease with an increased number of users. Thus, even though the greenbelt might have been able to provide a close to optimal level of ecosystem and recreational services at its inception, it is unlikely that it will keep doing so following significant population and economic growth (Lee and Linneman, 1998).

← 2. For example, landowners in German and Korean urban areas were not compensated for the loss of their development rights when their properties were included in a greenbelt (Lee, 1999; Siedentop et al., 2016). In 1998, however, about 21 years after the establishment of the first greenbelt in Korea, a Constitutional Court decision ruled that those who lost development rights due to the greenbelt regulation should be compensated (Bengston and Youn, 2006; Cho, 2002).

← 3. Brueckner and Kim’s (2003) analysis of stylised monocentric city models shows that the direction of the effect of property taxes on urban density inter alia depends on the magnitude of the elasticity of substitution between housing and other goods. A higher elasticity implies a lower likelihood of property taxes encouraging low-density development.

← 4. Empirical work in this area has been limited to the context of the United States and is inconclusive. On the one hand, the cross-sectional analysis of Song and Zenou (2006) shows that the average statutory propertytax rate has a negative effect on the size of the urban area, for a given population size. On the other hand, the panel data analysis of effective property tax rates of Wassmer (2016) reveals an effect in the opposite direction.

← 5. The effectiveness of split-rate property taxes in increasing density is also supported by empirical evidence on the United States. In their study of urban areas in Pennsylvania, Banzhaf and Lavery (2010) show that split-rate property taxation leads to higher urban densities, both in terms of structures and (to a lesser and more uncertain extent) population.

← 6. The majority of TDR programmes are established in areas with maximum density limits.

← 7. TDR programmes have been popular in the United States but have recently been implemented also in other OECD countries, including Australia, Italy and, although in a different context, the Netherlands (Harman, Pruetz, and Houston, 2015; Janssen-Jansen, 2008; Micelli, 2002).

← 8. In practice, this might be a challenging task, as it implies that policymakers have complete information about the external benefits of open space conservation and the external costs of development (Mills, 1980).

← 9. The demand for TDRs is in the majority of cases lower than that anticipated by planning agencies (McConnell and Walls, 2009).

← 10. Market activity can be facilitated and transaction costs can be reduced with the help of TDR banks, i.e. entities officially accredited by local authorities to purchase and resell TDRs. TDR banks can further stabilise the market and alleviate sellers’ concerns over finding purchasers for their TDRs in the shortterm (Pruetz and Standridge, 2008).

← 11. Empirical evidence of the effects of road pricing schemes or parking policy reforms (see relevant section below) on urban form are generally absent from the literature. An important reason for this is probably that the majority of road pricing schemes and parking policy reforms were introduced rather recently, therefore not allowing an empirical investigation of their effects on long-term changes in urban development patterns.

← 12. As already discussed, road pricing is a more efficient instrument to price the external costs of air emissions from car use.

← 13. The impact of on-street parking underpricing on the total private costs of trips is uncertain, as it depends on the time and other costs incurred in cruising to find a parking space.

← 14. This is also in line with the three pillars of the Aarhus convention on access to information, public participation in decision-making and justice in environmental matters (UNECE, 1998).