Abstract
China is seeking to prohibit the construction of any new gated communities and to gradually open existing schemes after three decades of growth of large-block gated estates. In this article, we use permeability analysis to explore the ‘what if?’ question posed by the policy: what if gated communities became permeable? We ask the question in respect of non-motorised access. We use two permeability metrics, closeness and betweenness, as outcome measures of gated and non-gated versions of the city. We construct a bespoke complete pedestrian network, rather than using the road network, for our permeability modelling. Nanchang, a medium-sized Chinese city with widespread gated communities, is our study area. A series of permeability analyses with and without gated communities is conducted using GIS and spatial design network analysis (sDNA). On the basis of these analyses, we sequentially sort the gated compounds whose opening will maximise permeability gains with minimum expropriation of property rights through coercive ungating. We offer the analysis to urban scholars, planners and governments by way of a quantified simulation. This study and methodology, which is transferable without high data requirements, can assist urban practitioners in reconfiguring urban form to promote a healthier living environment (more walking) and more economically viable local service centres (greater pedestrian footfall concentrations).
Introduction
Gated community research has tended to focus on social cost-benefit calculus, with an emphasis on radical social theoretic commentary in the 1990s giving way to more pragmatic social commentary and a growing slate of empirical studies measuring alleged social and private costs and benefits (Le Goix, 2005; Low, 2001; Webster, 2001). Much less effort has been given to normative design studies of gated estates or to empirically based theories of the physical urban form associated with gated communities (Miao, 2003). Urban planners must understand not only the functions and processes that create urban patterns, but also the direction in which urban form should change under various normative drivers (Talen, 2002). Social science studies of urban planning should deliver something back to practitioners.
Private neighbourhoods with their gates, impermeability and self-sufficient club economies have significant implications for cities. A good city form is about having permeable spatial patterns that can facilitate accessibility and maximise connectivity (Lynch, 1981; Talen, 2002). New urbanists argue that urban areas should strive for transit, pedestrian and bicycle systems access, while reducing dependence upon the automobile (Talen, 2013). Permeability, permitting ease of movement and avoiding urban severance, is generally considered a positive attribute of an urban design scheme (Melia, 2012). Gated communities, by definition and geometric and economic necessity, tend to segregate urban physical structure and to fragment urban spaces. Services, goods and facilities are provided exclusively to community members (Glasze et al., 2004; Miao, 2003). Access restriction is hard-wired into the business rationale of residential clubs. This means that the division between gated and ungated communities creates new socio-spatial patterns. The segregation raises important questions about the future forms of urban development, since a gated city tends to have sharper social geography divisions at a fine scale than traditional cities (Manzi and Smith-Bowers, 2005; Scott, 2002; Webster et al., 2018). One question arises: is it possible to maintain an open and permeable city while accommodating the seemingly unstoppable gated community genre of development, with its necessary degrees of enclosure? Answering this requires us to be specific about the particular performance domains of gated cities that might need to be controlled. In the study, we focus on a single dimension of urban performance, namely permeability. This allows us to answer the question: how would accessibility be re-distributed across the city if gates were removed? This, in turn, permits us to ask: how could we prioritise gated communities for ungating if we wanted to achieve maximum accessibility improvements with the least possible social and legal disruption?
China is seeking to prohibit construction of any new gated communities and to gradually open existing schemes after three decades of growth of large-block gated estates. New guidelines released by the central government in February 2016 call for more compact, livable, green and sustainable cities (Normile, 2016). This is a turning point in city form that breaks with the traditions of both planned and market-led economy periods. City form immediately post-1949 was heavily influenced by soviet-inspired wide streets, huge blocks and work-unit estates (danwei). US-style urban arterials and ringways were added from the late 1990s, and the land between became rapidly filled with commodity housing interspersed with commercial and industrial buildings, most often planned in impermeable superblocks. Virtually all commodity housing in China has restricted entrances, and a growing number of developments are in super-block gated compounds.
There is research that supports the central government’s wish to ungate China’s urban super-blocks, especially healthy city research that emphasises the importance of street connectivity for health (Chen et al., 2016; Day et al., 2013), under the assumption that urban form can affect physical activity, especially walking and cycling. Other studies show that more highly connected and permeable locations are associated with better physical and mental health, sustainability of transport, reduction of traffic congestion, improvement of air quality and economic wellbeing and inclusion of communities (Alfonzo et al., 2014). Physical inactivity, overweightness and obesity in car-oriented gated compounds with exposure to heavy air pollution in daily life is popularly understood to be a major risk factor in China. If urban planning is the biggest possible public health intervention, removal of the gates in Chinese cities could potentially have major health benefits through increasing walking and cycling. The literature suggests that this might significantly help lower the risk of the major non-communicable diseases such as coronary heart disease, type 2 diabetes, obesity and shortened life expectancy (Lee et al., 2012). We note, however, that the Chinese government’s interest in opening the gates is not based on health performance. The policy largely responds to the perceived negative effects on urban movement and congestion of traffic – large closed blocks are thought to be inefficient in this respect.
In this article, we use a permeability perspective to ask the question: ‘what if gated communities become permeable for non-motorised access?’. We propose two permeability metrics, closeness and betweenness, to use as outcome measures of gated and non-gated versions of the city. We construct a bespoke complete pedestrian network, rather than the traditionally-used road network, for the permeability modelling. Nanchang, a medium-sized Chinese city with widespread gated communities, is our study area. A series of permeability analyses of the city, with/without gating, is conducted using a geographic information system (GIS) and spatial design network analysis (sDNA) (Cooper et al., 2016). On the basis of these analyses, we sequentially sort the gated compounds whose ungating will create maximum effects. This allows a kind of mini-max analysis of the issue: which gates to remove to maximise permeability gains with minimum expropriation of property rights through coercive ungating?
Gated communities and car-oriented urbanism
Physical layouts of gated communities in China
Gated community in contemporary China presents a tangled picture in terms of development process and physical, economic and social outcomes. Two main types of gated compound layouts coexist: legacy danwei estates; and privately developed, secured gated compounds.
Although there was wide variation among the soviet-style work-unit compounds, in terms of scale, facilities and social services, there was an overwhelmingly ubiquitous presence of walls and guarded gates. Facilities in danwei were exclusively for the use of unit members. Through-traffic was forbidden at any time. From a bird’s eye view at the city scale, large urban compounds consisting of several to tens of acres are typical, distinguished by the regularity of the block-house layouts and unified building heights of five to six walk-up floors, with walkways laid out between the blocks and only minimal connections with local streets outside the compound (Glasze et al., 2004; Wu, 2005).
Contemporary Chinese gated compounds are usually large, involving hundreds of acres and thousands of residents. The built forms tend to be high-rise arranged in novel layouts, with much of the ground area devoted to elaborate landscaping and having clearly identifiable boundaries and gated entrance ways. Their physical manifestations include natural and human-made walls, or outward-facing buildings constructed around the periphery, all of which have the effect of physically segregating gated communities from surrounding areas. Some work units belong to state-owned enterprises that are still viable or which have diversified their enterprises and have renovated their gated danwei properties to provide employees with updated living conditions (Lu, 2006). Many modern gated communities have been built on inner-city industrial danwei areas as brownfield redevelopments. Others have been built on converted farmland as the city has expanded.
Car-oriented urbanism and gated compounds
Legacy danwei systems still impact city structure and travel behaviour (Sun et al., 2016). Most danwei gates nowadays are open in the daytime and provide free access for outsiders by foot or bicycle. Some are closed at night for security reasons. These gates do not bother local people, who are familiar with them and their opening hours. However, they are barriers for strangers, inhibiting through-travel (Xu and Yang, 2009). The road system of work units has never been integrated into the general road network of the city. Whereas local distributor roads in a western city will have been designed to roughly optimise local and trunk accessibility for local residents and other users, local distribution networks within a danwei are more likely to be determined by the gaps between buildings and internal circulation logic, with no reference to the next level of network outside.
Similarly, streets internal to contemporary gated compounds are also isolated from city road networks. In the immediately surrounding areas of gated communities, there are often extensive lengths of ‘dead-frontage’ building setbacks abutting pavements and devoid of retail and amenities. This is especially the case in the new development districts on the fringe of cities all over China. This tends to give rise to limited local street vitality and an unstimulating experience for walking, and is thus associated with lower non-motorised rates of movement (Day et al., 2013). Travelling between the superblocks of these communities involves walking long distances because of the lack of permeability. Pedestrian and cycle trips necessarily have to be routed outside of the gated blocks and onto arterials that carry automobiles, trucks and buses in growing numbers (Tao et al., 2010). Clustered gated compounds, with impermeability and mega-block structures, are likely to induce greater car-driving behaviour (Normile, 2016).
Permeability and its measures
Permeability has two important functional characteristics in urban design analytics. First, it permits pedestrians and cyclists from the wider area to access facilities, goods and services within a neighbourhood centre. This can be described as to-movement. Second, pedestrians and cyclists can travel through the neighbourhood directly, rather than detouring outside the neighbourhood, to reach other communities beyond. This can be described as through-movement. Many legacy danwei estates in Chinese cities allow non-motorised access to the community’s internal facilities, meaning that they are permeable in terms of pedestrian to-movement. However, where there are gatekeepers and a well-organised property management company, pedestrian through-travel may be restricted. Danwei communities therefore can be classified as urban areas with mixed and unstable permeability.
Allowing to-movement and through-movement for walking and cycling are two characteristics of a neighbourhood with good permeability. In this study we wanted to measure permeability after the hypothetical removal of gated community barriers, as demanded by central government policy, and to compare it with permeability with the current level of gatedness. To measure permeability, we borrow theories from network analysis (Newman, 2010).
Closeness: Measuring permeable to-movement
For the to-movement component of permeability we use closeness centrality. Closeness is commonly used in social and other network studies. It measures the mean distance from a node to other nodes using the shortest path. We use Euclidean distance to measure the shortest path. It takes low values for nodes that are separated from others by only a short geodesic distance. Such nodes in an urban street network are on average likely to have better access to traffic, services and information, being more directly connected to and influenced by other nodes. The higher a node’s closeness, the more central it is in relation to all other nodes within the part of the network included in the centrality analysis.
Mathematically, we denote
Betweenness: Measuring permeable through-movement
For the through-movement component of permeability we use betweenness centrality. Betweenness measures the density of shortest paths between all nodes in the network (within a given radius) passing through a link. It measures an important idea regarding through-movement in network analysis, that of the flow of information or other traffic and of the influence that vertices might have over that flow. Nodes with highest betweenness are the ones most likely to be the location for activities necessitating communications between a large number of people and businesses because they lie on the largest number of paths taken by traffic flows, urban goods and information. These places have the potential to attract more traffic and more people onto the street, and thus predict the location of commercial centres of varying scales. The radius at which betweenness is measured can indicate the type of traffic likely to pass through a node or link with a high betweenness value. Measured within a walking radius of 400 metres, for example, a high betweenness value can be taken as an indication of high pedestrian through movement and is likely to be correlated with locally-oriented commercial land uses.
Mathematically, the betweenness centrality of a node (place)
Geographic extent of permeability for walking and cycling
As implied above, we need a reference point for empirical studies of behaviour to interpret permeability measured by network metrics. Following published studies, we use 400 metres to represent an easy walking radius and the local scale for our analysis. We use 1600 metres (around 1 mile) as the longest walking distance. Empirical studies in Chinese cities have found that people would walk longer if the walking environment was supportive (Sun et al., 2016). 1600 metres is also an easy urban cycling distance. Finally, 5000 metres is regarded as the regular tolerable cycling distance in China and we use it to represent cycling distance and the regional scale for our analysis (Zacharias, 2005).
Construction of an urban pedestrian network model
Road centrelines tend to be used as a proxy for pedestrian routes in walking studies, justified on the grounds that most footpaths are pavements along roads (at least in North America and Australia). However, road networks often do not, in fact, reflect the finer-grained paths governing non-motorised access (Iacono et al., 2010). A study in Perth, Australia, dramatically found that measures of connectivity for some neighbourhoods improved by up to 120% when using pedestrian networks compared with using road networks (Chin et al., 2008). In a study of Davis, California, the pedestrian network within a five-minute walk of households was underestimated by as much as 40% when using street network as a proxy (Tal and Handy, 2012).
There are few studies of pedestrian networks in Asian cities, where urban design principles and density are strikingly different from western counterparts. The typically missing pedestrian network elements from official road network maps in China include the following: (a) Small alleys, which provide fine-grained permeability in the built environment and enhance non-motorised access; (b) Informal footpaths, which are routes not official or legally sanctioned but frequently used by local residents; and (c) Open accesses, which are those gates and pedestrian paths in danwei compounds familiar to local residents but unfamiliar to strangers. In addition, the large number of gated estates means that a significant part of the pedestrian network of a Chinese city is not included in the road network. Therefore, the official road centreline map would clearly be inappropriate for our current study as gated-ungated differences would not be captured.
Using high resolution aerial photographs, street views and field visits with GPS ground truthing techniques, we constructed a bespoke pedestrian network of Nanchang city, covering an area of 350 km2, a linear network of 5572 km and comprising 85,857 road segments (links between junctions). The object was to create what we believe to be the first city-wide pedestrian network map for a Chinese city and to use it to conduct the comparison of permeability and accessibility with and without gates. This entailed mapping a progressive three-tier pedestrian network: (i) pedestrian network along the official road network, referred to in our discussion as the official pedestrian network; (ii) official pedestrian network plus those missing elements listed above, referred to as gated pedestrian network; and (iii) pedestrian network including that part of the network within closed gated communities, which we refer to as the ungated pedestrian network.
Empirical study
In the empirical study, we develop a methodology that can be used for evaluating urban planning and design options, assessing the permeability of gated cities and conducting ‘what if?’ studies of the re-engineering proposals. We use heuristic network analysis metrics as surrogates for actual pedestrian and motorised traffic flows, making this a highly accessible technique suitable for rapid design appraisal and ongoing monitoring (Hajrasouliha and Yin, 2015). Our performance measures capture two essential features of the city with and without gates: (a) the network’s ability to make urban places accessible for non-motorised travel (closeness); and (b) the network’s ability to sustain the vitality of streets by allowing through-travel of pedestrians and cyclists (betweenness). Using the before-after performance results, we sequentially sort the gated communities to distribute the ungating intervention across the city in a way that will create maximum gains in permeability with minimum destruction of existing property rights.
Data and materials
Nanchang is a medium-sized Chinese city and the capital of Jiangxi province. In 2014, Nanchang’s urban population was 2.4 million and is projected to become 5.3 million by 2030 (Nanchang Statistics Bureau, 2015). The legacy of danwei is well preserved in the city, with large sized compounds built by former and surviving state-owned enterprises. Private secured neighbourhoods have developed apace in the city’s new development districts and suburbs. Nanchang is divided by the Gan River. The traditional city area, located on the eastern side of the river, has dense residential areas, commercial centres and public facilities. Development of the western side of the river started in 2000, under a new district development plan, resulting in a mix of high-density office buildings, residential communities and a network of large avenues. The city is experiencing rapid motorisation, bringing increasingly severe congestion to the city core (Qu et al., 2014).
Distribution of gated communities in Nanchang
To get a better understanding of how the city is segregated by gated compounds, we firstly identified all gated communities, including danweis and private secured residential compounds. We used Chinese map source services, including Baidu (map.baidu.com), Gaode (gaode.com) and Tengxun (map.qq.com), to identify and map gated communities. We also used high resolution aerial imagery and street views of the map sources, and complemented these with field visits.
Usually, Danwei compounds can be identified by their legacy names, such as a shortened company name. For example, the ‘Jiangfang Residential Five’ is associated with the Jiangxi Textile Factory and ‘Five’ is one of a large area of numbered residential areas for this danwei. The Hongdu of ‘Hongdu Residential Second’ was built for workers in the Jiangxi Hongdu Aviation Industry Group Corporation Limited. Other strategies for mapping gated danwei estates included identifying danwei hospitals that are still functioning (hospital services like most other urban services were provided by large work units) and primary schools that are still using the name of the danwei that built them for children of their workers. On the other hand, contemporary gated compounds can usually be identified by their western style names using modern terms, for instance: Honggu Triumphant Xiaoqu, Weidong Garden Xiaoqu and Fenghe New Town; where Honggu, Weidong and Fenghe are location names of districts, former villages or large urban avenues. Regardless of the names, both danwei and contemporary gated compounds are recognisable from aerial imageries by distinctive geometries of physical layouts.
Using these methods, we identified all the gated residential communities in the city. In total, there are 159 remaining danwei communities and 417 privately secured gated communities. On average, contemporary gated compounds are over twice the area of danwei in geographic size (7.13 versus 3.27 hectares). The territorial and service scale of contemporary gated compounds varies project by project, from just two or three apartment buildings in a small plot aimed at specialised end-users, to a large residential district of over 130 hectares, comprising high-, mid- and low-rise building types and targeting clients of mixed income levels.
Figure 1 shows the distribution of Nanchang’s gated communities. Danwei estates are mostly in the traditional city centre since most were built before the mid-1990s. A few are close to the present suburbs, such as Hongdu residential areas and Jiangfang residential areas which were located on the suburb’s fringe before recent urban expansion. The danweis close to the suburbs generally remain intact, with very few redevelopments, and are usually fully open, without gates or walls and with full non-filtered permeability. In the city centre danwei estates retain minimal gates and well-preserved walls. Privately secured communities dominate the new development district and most of the suburbs. Most contemporary gated communities were initially located on newly developing fringe areas, but many have become integrated within the city centre due to radical expansion of the urban area.
Distribution of the gated residential communities in Nanchang, China.
Three-tier pedestrian networks
As described above, we constructed a three-tier pedestrian network of Nanchang. Using network segments (links) as a measure of network size, the ungated pedestrian network is 18.3 times the size of the official pedestrian network (N = 85,857 compared to N = 4699). Using road intersections, a measure of connectivity, the ungated pedestrian network is 16.8 times the size of the official network. The mean length of network segments in the official pedestrian network is 305 metres, compared to 65 metres in the ungated pedestrian network. If Nanchang’s gates were opened, there would be a 138% increase in total pedestrian network length, with an extra 31,731 road segments and 17,974 road intersections, and average link size would reduce by 79%.
Analysis
We deepen this initial analysis by comparing the networks using systemic performance measurements associated with network functionality. We firstly calculate closeness and betweenness measures for long-walking but easy bicycle distance, using a 1600-metre radius. From the results, we select two areas for a detailed comparison with and without gates. Secondly, we calculate betweenness and closeness based on the two tiers of network: gated pedestrian network and ungated pedestrian network. The two non-motorised travel standards of 400- and 5000-metre radius as local and regional scales are applied for the ungated/gated scenarios of the whole city separately. Thirdly, we examine the change in betweenness and closeness when gates are ‘opened’. Pearson’s correlations, in quintiles of these changes, are calculated. We identify the top 20 percentage of road segments in terms of change in betweenness and closeness and make a line density analysis, weighted by betweenness or closeness change. We then identify the gated communities whose opening will potentially have the minimum destruction but approach the maximum effects.
We used GIS (ArcGIS 10.3, ESRI, Redlands, California, USA) to calculate and visualise the effects of making the city permeable for walking and cycling. sDNA, a plugin for GIS software, was used to calculate betweenness and closeness measures of connectedness. sDNA has been validated in the literature and in empirical studies of non-motorised travel behaviour (Cooper et al., 2016).
Results
Permeability measured by betweenness and closeness
We use a large residential development in the suburbs as our first case to illustrate changes in closeness with and without gated compounds. The distance from the area to the city centre is around 12 km. It contains 11 adjacent residential communities, of between 30 to 80 hectares each. Green Mountain Villa, for example, is around 80 hectares, while the Nanchang Residential ‘Theme Park’ (juzhu zhutigongyuan) has three sub-communities each of around 33 hectares. There are very few bus lines from this large residential compound connecting to the job centres of the city. Residents heavily rely on cars. There are few services and few workplaces. The large compound units are gated. Navigating to destinations within this large residential area mostly requires travelling along arterials and community walls, before entering the gates for local navigation.
Figure 2-C1 shows the current pattern of connectedness (closeness) and Figure 2-C2 the pattern after removing the gates to make the neighbourhood permeable for non-motorised travels. There is a striking difference, with a denser and more connected pedestrian network after the gates have been opened and with many more streets having high accessibility values. From this analysis, it is likely that at least two new community centres would emerge without the gates, shown by the streets with a much redder colour in Figure 2-C2. If a planning objective is to diversify the city and make use of the public benefits of streets in distributing economic choice and agglomeration benefits, then these two under-used parts of the city’s street network would be good candidates for the zoning of services centres and workplaces.
Closeness and betweenness measures for the permeability: A comparison before and after open the gates for pedestrians and cyclists in two selected study sites.
We use a second case, located five minutes’ walking distance from the city centre, to illustrate changes in betweenness with and without gates. This comprises large parcels of land with a mixture of legacy danwei and contemporary residential developments. It includes danwei such as the Nanchang Diesel Engine Factory Residential Areas and the Jiangxi Xinhua Print House Residential Areas, and contemporary gated compounds such as the Hengmao International Grand Garden. It is striking for its juxtaposition of gated commodity housing and danwei by land redevelopment.
Betweenness was computed to measure changes of through movement with and without gates. Comparing Figure 2-B1 and Figure 2-B2, the increase in potential for through-travel is mapped. Several pedestrian road segments start to play important roles in the simulated ungated city centre. One way of interpreting these analytical maps is that they quantify the amount of public value that is removed from the public domain as a result of gating. After ungating, through-movement increases on nearly all road segments in the neighbourhood, effectively extending the city centre’s commercially viable area. As well as increasing the number of streets in the city centre with high potential commercial vitality, the figure shows why opening the gates would attract non-motorised travel from roads that are currently suffering the most severe congestion in the city.
Changes in permeability after opening gated communities
Pearson’s correlation coefficients for a comparison in quintiles between ungated and gated scenarios are shown in Table 1. We found that the highest correlations of the ungated/gated are in the top 20% percentages (correlation coefficients are above 0.88, excluding closeness in 400 metres’ radius). This means that the accessibility of places with high betweenness or closeness when gated is consistently enhanced by the opening of the gated communities. The other percentiles of betweenness and closeness change inconsistently between ungated/gated scenarios. The inconsistent effect of ungating on these parts of the networks may be attributed to variations in the size of the surrounding gated communities, the geographic location of the road segments within the city grid (e.g. in the centre or on the fringe) and the heterogeneity in pedestrian network layouts within the calculation radius. As a result of these morphological variations across gated communities, opening the gates affects the urban grid connectivity and the associated positive and negative externalities to varying extents in different parts of the city. This leads us to sort gated communities in order of the magnitude of the impact of ungating (see next section).
Pearson’s correlation of changes in gated/ungated in quintiles (N road segment = 53,411).
To visualise the city-wide effects of gating on urban centrality patterns, maps of betweenness scores based on the gated/ungated pedestrian networks were produced at both the local and regional scale (Figure 3). Using a 400-metre radius, we obtained betweenness of each road segment at the local scale. The red parts of the network are those that are well configured for walking. The most highly permeable places post-gates are mostly scattered in the traditional city area. The maps also reveal a few candidate local centres on the west bank of the river and in the suburbs. These cluster in an originally free-standing town before urban expansion, as well as in the southern part of the city, which is now a large residential area. The maps identify candidate places where new local centres could be located as a policy intervention to densify and diversify walkable suburbs (Chiaradia et al., 2012). In the second group of maps, with betweenness measured within a 5000-metre radius, we see a clear enlargement of the vitality potential of the city following ungating. Notably, greater pedestrian and cycling interaction appears between the two riversides. The importance of the three bridges is revealed. Permeable bridges, only allowing pedestrians and cyclists but restricting cars, could greatly enhance such connections.
The changes in betweenness after opening the gated communities.
How could we select gated communities for ungating to maximise public benefits at minimum private cost?
There are high political and potentially legal and compensation costs in removing gates, so it is prudent to ask which ungating interventions might yield the greatest benefits. We generated a series of impact surfaces using the top 20% percentage changes in permeability (Chiaradia et al., 2012). The calculations are made separately at the local (400-metre radius) and regional (5000-metre radius) scales. We overlayed the surface with the distribution maps of the gated communities and then sorted the gated communities whose opening will bring the greatest permeability enhancements to the city.
Figure 4 and Table 2 show the results, indicating a partial ungating plan that optimises public gains for a given private loss. We found that the enhancements of betweenness through ungating are predominantly located in the traditional city areas, both at local and regional scales of analysis. The gated communities within the darker blue surfaces are where most danwei legacy estates are located. Our analytical prescription suggests that ungating danwei can be a priority. The opening of gated danwei estates might also have the advantage of facing less resistance, since many danwei communities allow mixed or unstable access, as already described. As well as opening gates, physical remodelling of these estates might focus on urban design improvements consistent with adding the internal grids to the city-wide grid to allow non-motorised through-travel.
The gates to remove to maximise permeability gains at minimum social cost.
A descriptive statistic of candidate communities for ungating.
Note: aThese high, medium and low ranges refer to the results in Figure 4.
Enhancements of closeness differ between local and regional scale radii. Closeness centrality is difficult to improve at the local scale merely by removing gates where the length of the gated communities’ walls is longer than 400 metres, which makes most destinations far away. At the regional scale, opening residential gated clubs in the new district and large residential areas on the city fringes would create more significant increases in permeability for non-motorised access and better access to destinations.
A combination of our maps provides useful information for calibrating an ungating policy. The selected (high potential accessibility change) gated compounds could, for example, be opened to non-motorised travel at the same time as removing some of the pedestrian/cycle through-routes from busier, more polluted and more dangerous routes around the edges of gated large blocks. This could also be accompanied by a number of specific and fine-grained design implications (Melia, 2012): a filtered permeable road network makes it easier to move by foot or on bike instead of in a car; while non-filtered permeable roads allow cars, but with traffic calming techniques, can slow traffic speed.
Discussion and conclusion
The release of new guidelines to open and prohibit gated communities is an interesting twist in China’s urbanisation story. Now the government would like to catastrophically re-engineer the morphology that has evolved under the gated development regime of the recent past. Whatever the private and political costs in terms of property rights appropriation, this is likely to have significant positive effects on spreading the urban local vitality and commercial potential that comes with pedestrian footfall and on health benefits, through a dramatic change in active travel behaviour.
In this article, we explore the ‘what if?’ question posed by this imminent natural experiment. What if we open the gates? We pose this as an intellectual thought experiment, recognising the difficulties and dilemmas in realising this policy. We do not take a position for or against the opening of gates. We merely ask how the movement potential of a heavily gated city would change if gates were opened. We offer the analysis to urban scholars, planners and Chinese governments by way of a quantified simulation.
Specifically, we use a permeability perspective to evaluate the through-movement and to-movement improvements achieved by ungating. Permeability is measured using closeness and betweenness centrality measures of a city-wide pedestrian network model. Using an empirical study of Nanchang, these two measures captured accessibility effects for pedestrians and cyclists, before and after the simulated ungating. This gives an indicator of the total urban accessibility benefits, or pedestrian route externalities foregone, as a result of closed-gated compounds. Alternatively, it can be taken as an indication of the social costs of gating.
We explored the potential of the city without gated compounds to: 1) bring forth new community centres; 2) promote the vitality of the city; and 3) provide guidance for an incremental ungating policy. We explored intra-urban variations in permeability, asking how much overall variation in permeability there is. We also explored changes in permeability in ungated/gated scenarios, asking how it is distributed across the city and which parts of the city can be better improved by ungating. Further, we sequentially sorted the gated communities, identifying those that will yield the biggest changes for a given private cost if opened up – those associated with the top 20 percent of enhancements in the city’s betweenness and closeness gains from ungating.
Our permeability analysis was based on a detailed pedestrian network model. As far as we know, this is the first urban planning policy ‘what if?’ study using a pedestrian network model for an entire city. The benefit of this approach can be seen in the striking differences between the three tiers of the pedestrian network, with an approximate 20-fold increase in road segments and intersections in the ungated network compared to the official road network, which would be the typically used network for pedestrian-level analysis at the urban scale. We offer this as a strong argument for building bespoke and accurate pedestrian network models as the basis for urban analysis of active travel issues and policies.
The re-engineering of an urban form dominated by gated communities faces massive challenges. It is not easy for current residents to accept ungating if no detailed plans accompany it, for instance to ensure security and ongoing service provision plans at the smaller neighbourhood or building scale. Developers are likely to take a cautious ‘sitting on the fence’ attitude with the policy, waiting for others to test out the government’s negotiating stance. However, we note that a policy that proscribes gated projects provides a new kind of mechanism for government planners to work more closely with developers to achieve good planning. A first result of the policy is likely to be the zoning of residential land in relatively smaller blocks than before.
The study has limitations and strengths. We did not separate out the pedestrian network from the cycle network in our analysis. The cycling mode-share in Nanchang is very low and there are barely any cycling facilities in the city (Nanchang News, 2010). Nonetheless, cycle route planning and the placement of cycling facilities could be guided by our models with little additional effort. We used centrelines to represent the pedestrian network since this was sufficient for our permeability analysis, based as it is on systemic connectivity. A finer pedestrian network could comprise pavements on both sides of a motorised road. We applied only a crude cost-benefit analysis to the normative part of this study. On the benefit side, we could have calculated net benefits by taking into account the reduction in monetary value of a residential location as a result of ungating, using hedonic modelling. Our model is a simplification of reality, but we suggest that it presents a powerful and policy-relevant analysis of the change in distribution of the city’s accessibility as a result of gates being taken down. The study is limited by having been conducted in only one city. However, we are confident that our methodology can be reliably adapted to other cities across China and elsewhere, with additional considerations, if needed, for local context.
The study has several notable strengths. It is the first study using a quantified simulation to test such a dramatic government policy reversal on gating. By quantifying the effects of opening gated compounds in pursuit of less congested and more accessible, healthier urban form in China, we both give scientific justification to the policy (on the benefits side) and at the same time provide specific policy design guidance on which gated estates to ungate. We offer a methodology for Chinese planners to conduct ‘what if?’ evaluations as they approach this massive central policy directive. We have suggested how a subset of gated compounds could be selected to maximise social gains for a given order of private costs. Finer, more pragmatic and sequential plans can be made on the basis of such strategic analysis. Assuming the ungating policy is widely implemented, we plan to measure the actual effects of this massive natural experiment, including economic, health and traffic effects over time.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
This study is supported by NSFC (41561035) and the 35th round of PDF/RAP Fellowship from University of Hong Kong.