Gas Leaks,Gas Shutoffs,and Environmental Justice in New York City

Abstract

Gas leaks in cities with older infrastructure are relatively common. The unburned methane which they release is a potent greenhouse gas with harmful health effects. Using administrative municipal data on gas leak reports, we provide a systematic analysis of residential gas leaks in New York City and their association with socioeconomic inequality. We find that both the reporting of gas leaks and the prevalence of resulting residential gas shutoffs is strongly structured by already existing inequalities across neighborhoods. Therefore, we argue that the gas infrastructure in urban areas is an important environmental justice issue as those communities who experience the brunt of failing gas infrastructure are the same communities who have faced decades of disinvestment and environmental racism.

Keywords

gas leaks environmental justice natural gas urban infrastructure utility shutoffs

Introduction

Unburned natural gas is a very potent greenhouse gas having a global warming potential that is 86 times higher than carbon dioxide over the first twenty-year period it reaches the atmosphere (Allen et al. 2014). Methane leaks from natural gas extraction and transmission are the largest source of anthropogenic volatile organic compound (VOC) emissions (West et al. 2006). In urban areas, VOCs react with other air pollutants, such as nitrogen oxides (NO_x), to form ozone (O₃), which is associated with premature mortality and can worsen the symptoms of asthma and bronchitis (Anderson et al. 2004; West et al. 2006).

Apart from contributing to climate change and air pollution, gas leaks present severe danger to people and property. Nationally, since 2002 – the earliest year for which systematic records are available – there have been almost 13,000 gas pipeline incidents which have caused 276 fatalities, 1,144 injuries, and more than $10 billion in property damage (PHMSA 2022). While most natural gas incidents occur in rural regions at the point of extraction and transmission, several high-profile explosions in urban areas have put in sharp focus the danger that the gas infrastructure poses to urban areas as well. For example, in 2010, an explosion of a natural gas pipeline in San Bruno, California, caused a fire which killed eight people and leveled an entire neighborhood. In 2020, a gas main explosion in South Philadelphia killed two people and destroyed five rowhouses. In New York City in 2014, a gas main explosion killed eight people, injured more than 50, and leveled two buildings in the East Harlem neighborhood of Manhattan (Santora 2014). A year later, an explosion due to negligent plumbing inside the basement of a building killed two people, injured more than 20 including firefighters, and destroyed three buildings in the East Village in Lower Manhattan (Solis 2015; Stempel 2017).

Deadly explosions in urban areas are relatively rare; however, gas leaks that do not reach sufficient levels to trigger an explosion are common especially in cities with extensive cast-iron gas main infrastructure (Jackson et al. 2014). Previous studies which have performed point-in-time measurements of street-level gas leaks in several cities with old infrastructure have found no relationship between the prevalence of gas leaks at the neighborhood level and neighborhood socioeconomic composition (Jackson et al. 2014; Phillips et al. 2013). In contrast, administrative data reported by utilities to the State of Massachusetts has revealed systematic inequalities in the density of leaks and the time-to-repair in communities of color and communities with lower-income renters (Di Gregorio and Jordan 2019; Luna and Nicholas 2022).

In this article, we provide a systematic analysis of residential gas leaks in New York City using administrative data from the New York City Fire Department (FDNY). We also examine an overlooked aspect of the consequences of gas leaks, namely the loss of gas service for cooking, heat, and hot water due to necessary repairs of the gas infrastructure. We seek to describe the demographic, socioeconomic, and housing characteristics of neighborhoods where the FDNY responds to reported gas leaks as well as the relationship between gas service shutoffs and socioeconomic inequality. Therefore, we expand the knowledge about an understudied urban environmental justice phenomenon. Our paper proceeds as follows. We first situate the study of the urban natural gas infrastructure within the literature on environmental justice. Second, we review the small but growing literature on the locations of urban natural gas leaks. We then describe the natural-gas infrastructure and its regulatory environment in New York City and present a new approach of studying urban gas leaks using administrative data produced by city agencies. Finally, we discuss how our results inform local climate policy and the push to electrify buildings.

Background

A large body of literature in environmental justice and urban political economy has described the proximity of low-income communities of color to undesirable land uses, the unequal provisions of public goods and services in such neighborhoods, and the unequally distributed power over decision-making processes in urban planning (Alesina, Baqir and Easterly 1999; Banzhaf, Ma and Timmins 2019; Logan and Molotch 1987; Mohai, Pellow and Roberts 2009; Taylor 2014). Low-income communities and communities of color are disproportionately exposed to air pollution (Ma 2020), water pollution (Schaider et al. 2019), traffic (Riedel et al. 2014), hazardous housing conditions (Weitzman et al. 2013), and hazardous waste facilities (Bullard 1990; Taylor 2014) while also having fewer neighborhood amenities such as parks and street trees (Schwarz et al. 2015). The environmental needs of such communities are routinely exacerbated by the unequal response of government agencies to environmental hazards (Konisky and Reenock 2018; Konisky and Schario 2010; Liang 2016), the unequal allocation of recovery funding (Elliott, Brown and Loughran 2020; Howell and Elliott 2018), and the unequal compliance with environmental regulations by businesses and utility providers (McDonald and Jones 2018).

The research studies above illustrate that environmental justice concerns broadly relate to problems associated with the unequal distribution of environmental burdens (also known as distributive justice) and the decision-making processes that lead to the unequal distribution of environmental burdens (also known as procedural justice) (Banzhaf, Ma and Timmins 2019). While natural gas, along with water and electricity, is a major component of the utility infrastructure of urban areas, it has received relatively little scholarly attention when it comes to its environmental justice impacts on urban areas from either a distributive-justice perspective or procedural-justice perspective. Part of this is because urban gas leaks have remained mostly low-grade and not easily detectable (Jackson et al. 2014; Luna and Nicholas 2022). Their cumulative impact on climate change, air pollution, and utility service shutoffs typically follows what Rob Nixon (2013) calls “slow violence” – or incremental and relatively invisible environmental catastrophes that fail to capture the public's attention, but in the aggregate and over time can be just as destructive as rapid-onset disasters.

Part of why urban gas leaks have received comparatively little attention is also because the switch from coal to natural gas in the residential electricity generation sector and from oil to natural gas in the residential heating sector has been hailed as a net improvement in terms of both particulate air pollution and greenhouse gas emissions (Ladage et al. 2021). However, there is a growing concern amongst climate scientists and environmental epidemiologists that the long-term contribution of gas to global warming might be on par with that of coal and that its use for heating and cooking can lead to adverse health outcomes due to its effects on air pollution (Howarth 2014; Willis et al. 2023).

In older cities in the United States, the distribution system of cast-iron gas mains dates back to the nineteenth century and is prone to high levels of leakage (Pevzner 2018). Despite efforts to decrease the mileage of leak-prone gas pipes, as of 2022 nationally there were more than 17,000 gas main miles and almost 6,000 gas service lines made out of cast iron (PHMSA 2022). Cities with leak-prone pipes have methane leaks of almost twenty-five times the volume compared to cities with pipes made from more modern materials (Gallagher et al. 2015; Jackson et al. 2014; Phillips et al. 2013; von Fischer et al. 2017). In New York City, as much as 2.2 percent of the natural gas in the local network is unaccounted (Center for an Urban Future 2014). Moreover, total gas leak volumes in the largest U.S. cities are systematically underestimated by the Environmental Protection Agency (EPA) as its inventory of fugitive methane does not include end-use emissions, such as gas that escapes inside buildings (Plant et al. 2019).

The failure rate to repair leaks in urban distribution networks is also high. A longitudinal analysis of leaks in Massachusetts shows that over 3 years about 20 percent of all gas leaks were either not repaired or needed additional repairs (Edwards et al. 2021). Since utilities usually prioritize the repair of more dangerous leaks and since states do not monitor repair success, there have been few incentives to take lower-grade leaks more seriously or to measure the climate impacts of failed repairs of leaks graded as non-hazardous in the immediate future (Edwards et al. 2021). Lack of data at small geographic scales has also hampered efforts to track and assess responses to gas leaks. In fact, Massachusetts is the only state in the U.S. that requires that utilities release detailed geographic data on the location of gas leaks, their severity, and their repair status (Luna and Nicholas 2022).

Moreover, gas leaks can also occur at several places in the urban distribution system. There could be a leak of the gas main pipeline which runs under city streets, a leak in the pipes that carry the gas from the mains to individual buildings (i.e., the gas service lines), a leak in the pipes in the common spaces and service rooms of apartment buildings, or a leak in individual housing units. A majority of street-level leaks come from the mains that run under each street, rather than service lines that feed into individual buildings (Gallagher et al. 2015). To our knowledge, there are no studies of gas leaks that occur due to faulty piping in individual buildings nor studies of gas shutoffs due to such faulty piping.

While there are no citywide studies of gas leaks inside individual buildings, there is a growing body of literature in engineering which has developed methods to measure the prevalence and precise geographic location of street-level gas leaks with methane sensors mounted on vehicles (Weller, Yang and von Fischer 2019). For example, a study of street gas leaks in Manhattan in 2014 measured 1,050 leaks per 247 road miles, or 4.25 leaks per mile (Gallagher et al. 2015). Only 11 of those leaks reached relatively high levels (Gallagher et al. 2015). The density of street-level gas leaks in Manhattan are of similar magnitude to the ones in other cities with old gas main infrastructure (Gallagher et al. 2015; Jackson et al. 2014; Phillips et al. 2013).

None of the studies using vehicle-mounted sensors to measure the occurrence of leaks have found a statistically significant association between gas leaks and neighborhood-level median income, poverty, and racial/ethnic composition (Jackson et al. 2014; Phillips et al. 2013). Nationally, gas utility service areas with greater proportions of owner-occupied homes and higher median housing prices have lower leak volumes, but there do not appear to be statistically significant associations between leak volumes and the median income, education, or racial and ethnic composition of these service areas (Scott, Scott and Greer 2019). However, research using administrative data of gas leaks reported by utilities to the State of Massachusetts over several years has found that both the exposure to gas leaks and the repair of gas leaks is highly correlated with neighborhood inequality, with people of color, households with limited English proficiency, renters, and lower-income households disproportionately being exposed not only to more leaks but also to longer periods of leak repair (Di Gregorio and Jordan 2019; Luna and Nicholas 2022). These findings suggest that even if gas leaks may be relatively widespread across urban areas with older infrastructure, their density and repair status over time are structured in a way that follows longstanding neighborhood racial and socioeconomic differences. Therefore, from a distributive justice perspective, there is some evidence that groups of people who have traditionally borne the brunt of other environmental burdens have higher exposures to urban gas leaks. From a procedural justice perspective, the experience of Massachusetts as the only state where utilities are required to report systematic data on urban leaks suggests that policy decisions around the repair of the natural gas infrastructure might lead to inequitable outcomes given the lack of information to support the regulatory processes of municipal and state actors.

The New York City Context

New York City has the second oldest gas infrastructure in the U.S. with some gas main pipes dating back to the second half of the nineteenth century when multiple competing utility companies vied for the local streetlight business (Pevzner 2018). As of 2021, NYC and its northern suburbs still had 1,989 miles of cast/wrought iron gas mains representing about 30 percent of the length of the local gas infrastructure (PHMSA 2022). According the 2021 American Housing Survey, 90 percent of households in the City use gas in any capacity; 74 percent of households use piped gas for cooking, 64 percent of households use piped gas as water heating fuel, 56 percent of households use piped gas as main house heating fuel (authors’ calculations).

While comprehensive plans to replace the City's cast iron gas mains date back to 2005 (Consolidated Edison Company of New York 2022), it was not until two deadly gas explosions in 2014 and 2015 when both New York State and New York City stepped up their treatment of gas leaks as a hazard to the public. In the aftermath of the first explosion, Con Ed entered into a $153 million settlement with the State and accelerated its timeline for replacing corrosion-prone cast iron pipes in its distribution network (Solis 2015; Stempel 2017). After the second explosion, the NYC City Council passed Local Law 152 which instituted routine periodic inspections of gas piping inside buildings and required that building owners immediately correct any leaks or non-compliant gas plumbing connections (NYC Office of the Mayor 2016). The local fire department also started responding to every report of a gas leak even when the report came in on the city's non-emergency line (Giambusso 2014).

What followed were lengthy reports in the local press regarding weeks-long gas outages in residential buildings, including many public housing properties (for some examples see Brand 2021; Miller 2016; von Oldershausen 2020). Residents lost the ability to use their gas stoves and needed to purchase either prepared food or shift to using electric appliances which increased the economic burden on low-income households who already pay disproportionate amounts of their income on utilities (Hernández 2013). Some buildings ended up without gas for almost a year while landlords, utility companies, and the New York City Department of Buildings passed the blame to each other regarding the length of gas outages (Burke and Gartland 2019; Small 2020). Tenants and housing advocates saw the lack of gas as another tactic of unscrupulous landlords to displace tenants so that they could raise rents (Whitford 2016).

As the gas infrastructure in New York City ages, it has become a cornerstone of local environmental justice efforts to transition buildings away from fossil fuels (Vardi 2021; Henchen and Kroh 2020; Walsh and Bloomberg 2023). Both New York City and New York State have now passed legislation that prohibits natural gas hookups in most new buildings. In 2019, New York State passed the Climate Leadership and Community Protection Act (CLCPA), which sets ambitious goals to reduce green-house-gas emissions by 2050 by 85 percent from their 1990 levels (Morris 2019). This target would require eliminating the use of natural gas for heating and powering buildings (New York State Research and Development Authority 2022). The CLCPA also established disadvantaged community designations, or census tracts that are mandated to receive at least 35 percent of the state's “overall benefits” from clean energy and energy efficiency programs.¹ About 55 percent of all census tracts designated as “disadvantaged” are in New York City, representing 44 percent of all NYC tracts (see Figure 1).

Figure 1.

New York City disadvantaged community (DAC) designations under the 2019 Climate Leadership and Community Protection Act (CLCPA).

Given the intersections between natural gas, climate change, and environmental justice, in this paper, we advance the literature on urban gas leaks by examining not only their geographic location but also the related process of gas service shutoffs. We extend the analysis of gas leaks to instances of such that occur within residential buildings. We also consider the implications of both gas leaks and gas utility shutoffs within the broader framework of distributive and procedural environmental justice and discuss how the reliance on fossil fuels for heating and cooking intersects with long-standing inequalities by race/ethnicity, income, and nativity that place disadvantaged populations in the most dangerous urban spaces.

Analytic Strategy

We use maps and statistical analysis to describe the neighborhood-level correlates of reported leaks and gas shutoffs. We rely on 911 and 311 reports in New York City to ascertain the location of reported gas leaks and gas shutoffs. In the United States, 911 is the universal number to call in case of an emergency that requires immediate assistance from the police, fire department, or ambulance. 311 is a non-emergency line where residents of many cities can request municipal services such as trash clean-up or register a complaint for a non-life-threatening situation, such as excessive noise or rodent infestation.

By providing a range of communication channels and by storing and publicly sharing the reports received, the 911 and 311 reporting systems allow municipal authorities to track and address concerns in a timely and transparent manner. In NYC, 311 handles centralized requests for services in over 180 languages and can be accessed 24 h a day via phone, via the web, or through a smartphone app (Legewie and Schaeffer 2016). Previous research which has used 311 reports has examined environmental issues ranging from the geography of noise reports (Legewie and Schaeffer 2016; Liu et al. 2019), to the frequency of urban flooding (Agonafir et al. 2022), and the presence of rats (Sánchez, Rios and Murray 2021). Complaint data has also served as an indicator of community engagement and political participation (Lerman and Weaver 2014; Levine and Gershenson 2014).

Research that has used 311 reports has had to deal with two limitations to such data. First, despite the considerable geographic and time precision of 311 data, the reports do not include any indication as to what the demographic or socioeconomic characteristics of the person who made the report are. Therefore, following previous research, we aggregate all 911 and 311 data up to the census-tract level which lets us examine the association between the volume of 911 and 311 calls and census-tract socioeconomic conditions as measured by the Decennial Census and the American Community Survey (ACS). While our analyses can describe the associations between gas leaks and gas shutoffs and the socioeconomic conditions of the neighborhoods where they occur, they cannot be interpreted at the individual level in terms of either the propensity of specific types of people to report gas leaks or the occurrence of shutoffs in specific apartment units.

Second, reports to 911 and 311 reflect two separate factors: 1) the conditions which produce the complaint and 2) the “contacting propensity” of the population (Minkoff 2016). Neighborhoods with older housing, more owner-occupied housing, more residents above the age of 65, and more households with children are more likely to contact 311, but the associations between neighborhood socioeconomic conditions and the number of 311 reports is weak, suggesting that socioeconomic resources at least at the aggregate neighborhood level are not a major obstacle to submitting 311 reports (Minkoff 2016). In fact, in 2021 New Yorkers made almost 22 million calls to 311, which averages to 2.46 calls per resident per year. About 2 million of those calls are to request services from the local government (Legewie and Schaeffer 2016). Surveys of 311 show that New Yorkers are widely aware of its existence and are generally satisfied with the services provided by the call center (Johnson 2010; NYC Office of Technology & Innovation 2022).

In our statistical models of gas leak reports and gas shutoff reports, we use a list of covariates that according to prior research are associated with higher volumes of 311 reports. Still, the two factors that generate 311 reports – objective environmental conditions and contacting propensity – cannot be completely separated from each other, as the same communities that have higher social capital (and thus are more likely to call 311 for government services) are also more likely to have the economic capital to purchase housing in neighborhoods with better amenities (Minkoff 2016). This is a limitation that our research shares with all previous literature that uses 311 reports. In the Results section below, we implement several sensitivity tests designed to address the limitation of using 311 reports to measure objective environmental conditions.

Data

The data for our analysis come from two main sources: the New York City Open Data Portal and the U.S. Census Bureau. From the NYC Open Data Portal, we downloaded administrative data on gas leaks reported to the NYC Fire Department (FDNY) and on gas shutoff complaints reported to the NYC Department of Housing Preservation and Development (HPD). We accessed 2000 Decennial Census Data and 2008–2012 five-year American Community Survey (ACS) data in constant 2010 census tract boundaries from the online portal of Social Explorer. We used the first set of data to geocode the locations of gas leak reports and of gas shutoff complaints in each census tract in NYC. We used the second set of data to generate measures of the demographic and socioeconomic composition of census tracts in NYC including changes over time in the characteristics of census tracts.

FDNY Dispatch Data. The FDNY dispatch data that we accessed through the NYC Open Data Portal covers the period from 2013 to 2018. It includes the longitude and latitude of alarm boxes throughout NYC. We geocoded each alarm box to its respective census tract. Because the FDNY dispatch data is not building-specific, we could not link it to the data on gas shutoffs to examine more directly the relationship between leaks and shutoffs.

In 2013, the FDNY responded to almost 13,000 gas leak reports. Between 2014 and 2018, the FDNY responded to about 22,000 gas leak reports per year. For the purposes of our analysis, we aggregated the number of reported gas leaks to which the FDNY dispatched a fire truck within each census tract in New York City for the entire period between 2013 and 2018, using the 2010 Decennial Census tract definitions. Apart from the lower number of reported gas leaks to which the FDNY responded in 2013, there does not appear to be any other time trend to the data. As reported in the press (Giambusso 2014), the higher volumes of FDNY response to gas leaks after 2014 appear to be driven by a change in policy after the deadly gas explosion in Harlem. Currently all gas leak calls to 311 are forwarded to 911 and the FDNY. The two local gas utilities also encourage residents to call 911 upon any smell of gas.²

Gas shutoff data. We downloaded 311 complaints about gas service shutoffs from the NYC Open Data Portal. The complaint data covers all available years in the data portal as of the date that we accessed the portal in July 2021. The data covers the period between August 2014 and June 2021. There were a total of 69,489 gas shutoff complaints with an average of about 10,000 complaints per year. The complaint data covered 17,155 unique buildings. In addition to counting the total number of gas shutoff complaints, we also counted the total number of gas shutoff violations issued by the NYC Department of Housing Preservation and Development (HPD) as well as the average time (in days) that gas shutoff complaints remained open. There were a total of 16,928 violations issued with about 2,500 violations per year. The median time that complaints remained open was 13 days with the top 10 percent of complaints remaining open for more than 48 days. The number of complaints, the number of violations, and the average time that complaints remained open did not exhibit a particular trend over time, so we analyze them aggregated for all years in the dataset. We only analyze non-duplicate gas shutoff complaints as indicated in the notes that HPD includes as part of the disposition of each complaint that it receives. In supplementary sensitivity analyses, we also analyze gas shutoffs that have been confirmed as such by an HPD inspector.

We used the Borough-Block-Lot (BBL) identifier of each complaint to find the residential building address of the gas shutoff and then geocoded each address to its respective 2010 census tract location using the Census Geocoder API.

Census data. Most of our demographic and socioeconomic data for the analysis comes from the American Community Survey (ACS) five-year 2008–2012 dataset. We used this dataset to generate variables on the demographic, socioeconomic, and housing composition of each census tract. We also downloaded 2000 Decennial Census data in 2010 census tract boundaries to generate indicators of changes to the socioeconomic composition of census tracts over time. We describe each variable in our analysis in more detail in the Methodology section below.

Methodology

Our analysis combines maps and statistical regressions to examine the relationship between gas leaks, social inequality, and environmental justice in New York City. We first present maps which show the geographic distributions of gas leak reports and gas shutoff complaints across NYC and then model these distributions across census tracts using regression analysis. Since the dependent variables in most of our models represent counts of either 911 or 311 calls, we use negative binomial regression for our statistical analyses. These types of regressions are the appropriate way to model count data which consist of positive integers and which has overdispersion, i.e., the variance of the dependent variable is larger than its mean (Long and Freeze 2005). We use Ordinary Least Squares regression to model the average number of days gas shutoff complaints remain open. All regression analyses exclude census tracts that have no population or where the entirety of the population is in group quarters, i.e., where the population lives in institutional settings, such as prisons, mental hospitals, and military barracks. Our analyses also exclude park areas and industrial areas.

We elect to analyze the number of 911 and 311 calls as opposed to a normalized quantity of these calls, such as the number of calls per person or per square mile. As Minkoff (2016) has argued, it is unclear what the number of potential 311 callers within each tract is. Moreover, the incidence of gas leaks depends on the length and type of gas pipes under each street and the pipe infrastructure within each building. While the length of gas mains is available at the utility service-zone level, there are no analogous data at the census tract level (Luna and Nicholas 2022). As gas leak reports are broadly a function of both the number of people in an area and features of the infrastructure of that area, there are multiple measures of tract characteristics related to the number of potential reports. Therefore, we directly control for the population size of each tract as an indicator of potential 911 and 311 callers, for the number of housing units in each tract as an indicator of building-level gas infrastructure, and for the area of each tract as an indicator of the length of gas mains running under each street. In addition, we control for each tract's median housing value, median rent, and the median year the housing units in the tract were built as older and cheaper housing could potentially have more deficient piping. Given reporting in the local press on the circumstances surrounding gas leaks, we also measure the number of public housing units in each tract. We expect that tracts with more public housing units will have more reports of gas leaks to 911 but not 311 reports of gas shutoffs, as up until 2022, public housing residents could not log complaints with 311 but had to use a system internal to the local housing authority (Turetsky 2022).

All regressions include spatial lags for the respective dependent variables in each model. Spatial lag models statistically correct for the non-independence of individual observations that are geographically close to one another. They are used to model spatial diffusion processes where the value of the dependent variable of interest for any tract might be systematically correlated to the values of the dependent variable for nearby tracts. In the case of gas leak reports, there are several reasons why both the propensity to report gas leaks and the underlying structural conditions of the gas infrastructure are correlated across nearby tracts. As previous research has shown, the propensity to call 911 or 311 varies by neighborhood socioeconomic composition (Minkoff 2016; Levine and Gershenson 2014; Legewie and Schaeffer 2016; Sánchez, Rios and Murray 2021). Given NYC's high degree of residential segregation, we might expect that the patterning of either 911 or 311 calls to be correlated across space as tracts with more homeowners and higher income populations are highly clustered at the city level. The underlying NYC gas infrastructure also varies across space. Older neighborhoods especially in Manhattan and Brooklyn typically have older gas pipes running under the city streets and inside buildings (Pevzner 2018).

In supplementary analyses presented in Table 1 in the Appendix, we formally test for the spatial autocorrelation of our dependent variables and calculate the global Moran's I statistic for each. All Moran's I values are positive and statistically different from 0, which indicates that tracts with more reported gas leaks and tracts with more gas shutoff complaints are spatially clustered around each other. Therefore, in each of our regression models, we include an additional term, ρ, which captures the effect of the proximity of tracts in space (Ward and Gleditsch 2008). We estimate ρ using a row-standardized spatial weights matrix that represents the inverse Euclidian distance between the geographic centers of census tracts. The values of the matrix decline with distance from each focal tract. The inclusion of the spatial lag parameter controls for the extent to which nearby tracts have similar levels of gas leak reports or gas shutoff complaints.

Our regression models also include a series of demographic and socioeconomic covariates which capture both potential infrastructural differences across neighborhoods and different propensities to call 311 to report local problems. First, we control for the percent of the population over the age of 65 and the percent of households with children as neighborhoods with greater percentages of these two types of populations are the most likely to call 311 (Minkoff 2016). Second, we include covariates which describe the racial composition of each tract, the percent of residents who are foreign born, the percent of residents who are renters, the educational attainment of tract residents above the age of 25, and the percent of residents in poverty. Poor communities and communities of color on average have worse infrastructure compared to more affluent neighborhoods and neighborhoods with greater percent White residents (Taylor 2014). Lower income tracts are also more likely to report hazardous conditions and building-code enforcement issues compared to higher income tracts (Kontokosta and Hong 2021). On the other hand, those who have a college degree, who are born in the US, who are homeowners, and who are less likely to live in poverty may be more likely to have greater experience with city bureaucracies. They are also more likely to call 311 (Cavallo, Lynch and Scull 2014; Levine and Gershenson 2014; Minkoff 2016).

In addition, we include covariates which indicate the percent vacant housing and percent of residents in the tract that have lived there for less than year. Along with the covariates on tract socioeconomic composition, such as percent in poverty and educational attainment, these are established indicators of neighborhood socioeconomic disadvantage and residential instability which could plausibly be linked to poor neighborhood infrastructure (Sampson, Raudenbush and Earls 1997; Sampson, Morenoff and Gannon-Rowley 2002).

Lastly, we construct a measure of gentrification for each census tract in NYC. We distinguish between three types of tracts: 1) tracts that could not gentrify, i.e., tracts that had median income in 2000 that was above the median for all NYC tracts; 2) tracts that could gentrify but did not do so between 2000 and 2010, i.e., tracts that had below-median income in 2000 and experienced changes in tract rent and tract percent of residents with a college degree that was below the median for all NYC tracts; and 3) tracts that gentrified, i.e., tracts that had below-median tract income in 2000 and experienced a change in tract rent and tract percent of residents with a college degree that was above the median for all NYC tracts. This is a standard measure in the literature on gentrification (Freeman 2005). It captures the process of “the replacement of low-income, inner-city working class residents by middle- or upper-class households” (Hammel and Wyly 1996). Given the high share of renters in NYC as well as the greater degree of accuracy of reporting of rents as opposed to homeowner housing values in census data, we follow the Furman Center approach to identifying gentrifying areas which considers only changes in rent values over time as opposed to both changes in rent values and changes in housing values (NYU Furman Center 2016).

Gas leak reports and gas shutoffs might pick up an underlying change in neighborhood population dynamics. Neighborhoods which have experienced the influx of more highly educated residents might also be more likely to have higher 311 call volumes as newcomers demand greater attention from city agencies. Some housing advocates have also suggested that lack of gas – and lack of repairs in general – work in favor of building owners as potential high turnover of renters allows landlords to increase the rent on each subsequent new tenant after the necessary repairs are made (Whitford 2016). Therefore, both areas that are gentrifying and ones that could potentially gentrify may experience a disruption in gas services due to population turnover and displacement pressures in the housing market.

Results

Geographic Distribution of Gas Leak Reports and Gas Shutoffs

We start the discussion of our results with a description of where New Yorkers call the fire department to report gas leaks and where New Yorkers report gas shutoffs (see Figure 2 and Figure 3).³ Gas leak reports to the FDNY are made from almost all census tracts in the city, but they are widely concentrated in Manhattan, the Bronx, Northern Brooklyn, Western Queens, and Northeastern Queens. Manhattan has the greatest number of tracts that are in the top five percentile of reported gas leaks (or more than 192 reports of gas leaks per tract). Most of the neighborhoods with high number of gas leaks reports are either places with high number of people of color, high number of residents in poverty, or both (see Figure 4). About half of all tracts that are in the top five percentile of the number of reported gas leaks are places with more than 500 public housing units. Such tracts represent only 5 percent of all tracts in New York City.

Figure 2.

Total number of reported gas leaks (2013-2018), New York City (census tracts).

Figure 3.

Total number of gas shutoff complaints (2014–2021), New York City (census tracts).

Figure 4.

Racial/ethnic and socioeconomic composition, New York City (ACS 2008-2012, census tracts).

The spatial distribution of gas service shutoff complaints largely follows the distribution of gas leak reports, with shutoff complaints concentrated in lower-income neighborhoods and neighborhoods with substantial presence of Black and Hispanic populations.⁴ Gas leak reports and gas shutoff complaints also largely follow the distribution of census tracts designated as “disadvantaged” under the 2019 CLCPA (see Figure 1). Therefore, the brunt of residential gas shutoffs, including confirmed shutoffs for which building owners have been issued a violation (see Appendix Figure 1), falls on lower-income communities of color.

Regression Analyses

Our tract-level regression analyses of the association between tract socioeconomic characteristics, gas leaks, and gas shutoffs confirms the relationships present in the data visualizations. Table 1 shows the means and standard deviations of all dependent and independent variables; Table 2 displays the regression results. On average, the FDNY responded to 59 gas leak reports per tract between 2014 and 2018, while residents logged 33 complaints of shutoffs per tract between 2014 and 2021. As suggested by the maps in Figure 2 and Figure 3, the number of gas leak reports and gas shutoff complaints as well as the number of gas shutoff violations issued by the HPD varies widely across neighborhoods with the standard deviation of each of these variables being higher than its mean.

Table 1.

Descriptive Statistics, NYC (Census Tracts).

	Mean	SD	N
Number of reported gas leaks (FDNY)^a	59	66	2,063
Number of gas shutoff complaints (311)	33	58	2,105
Number of gas shutoff violations, HPD (311)^b	9	14	1,922
Average time gas shutoff complaints remain open (311)^b	22	18	1,922
% non-Hispanic White	34	31	2,105
% non-Hispanic Black	24	30	2,105
% Hispanic	27	23	2,105
% non-Hispanic Asian	13	16	2,105
% non-Hispanic of other race	3	4	2,105
% with college degree (age 25+)	31	20	2,105
% foreign born	37	16	2,105
% below federal poverty level	19	13	2,105
Tract gentrified between 2000 and 2010	0.11	0.32	2,105
Tract did not gentrify between 2000 and 2010	0.39	0.49	2,105
Tract not gentrifiable, 2000	0.50	0.50	2,105
Total population	3,887	2,131	2,105
% age 65 or over	12	7	2,105
% moved into tract <1 year ago	9	6	2,105
% households with children	34	13	2,105
Total housing units	1,601	1,083	2,105
Tract area (square miles)	0.12	0.20	2,105
Median housing value	506,804	223,150	2,105
Median rent	1,237	344	2,105
% renter households	63	25	2,105
Median year housing built	1950	13	2,105
Percent vacant units	9	6	2,105
Number of public housing units	78	291	2,105

The analysis of FDNY dispatch data excludes all census tracts in the Highbridge/Concourse and Fordham/University Heights Community Districts of the Bronx as there were no FDNY dispatch data for this particular area of the city in the NYC Open Data portal.

The analysis of gas shutoff violations and the amount of time complaints remain open only includes census tracts with at least one gas shutoff complaint.

Table 2.

Regression Models of Gas Leak Reports to the FDNY and 311 Complaint Data.

	Number of reported gas leaks (FDNY dispatch data)	Number of gas shutoff complaints (311 data)	Number of gas shutoff violations (311 data)	Average time gas shutoff complaint remains open (311 data)
	negative binomial model	negative binomial model	negative binomial model	OLS model
Tract racial/ethnic composition (% White omitted)
% Black	0.008***	0.016***	0.010***	−0.006
	(0.001)	(0.002)	(0.001)	(0.020)
% Hispanic	0.005***	0.012***	0.008***	0.011
	(0.001)	(0.002)	(0.002)	(0.028)
% Asian	0.006***	−0.0002	−0.001	0.032
	(0.001)	(0.003)	(0.002)	(0.043)
% other race	0.002	0.016*	−0.001	0.036
	(0.004)	(0.007)	(0.006)	(0.094)
Tract socioeconomic composition
% with a college degree	0.004*	−0.0001	0.008**	0.045
	(0.002)	(0.006)	(0.003)	(0.049)
% foreign born	−0.009***	−0.003	0.002	−0.056
	(0.002)	(0.003)	(0.002)	(0.035)
% below federal poverty level	0.005*	0.009*	0.009**	0.001
	(0.002)	(0.004)	(0.003)	(0.053)
Tract gentrification (not eligible to gentrify omitted)
Tract gentrified between 2000 and 2010	0.043	0.177	0.197**	1.381
	(0.056)	(0.101)	(0.072)	(1.892)
Tract did not gentrify between 2000 and 2010	0.111*	0.263**	0.287***	0.518
	(0.052)	(0.099)	(0.074)	(1.439)
Tract demographic composition
Total population (in ‘000s)	0.156***	0.166***	0.011	0.458
	(0.025)	(0.040)	(0.035)	(0.482)
% age 65 or older	−0.001	−0.022**	−0.003	0.061
	(0.004)	(0.008)	(0.005)	(0.145)
% moved into tract <1 year ago	−0.002	−0.002	−0.0002	−0.073
	(0.004)	(0.005)	(0.005)	(0.086)
% households with children	−0.011***	−0.014**	−0.0004	−0.029
	(0.003)	(0.005)	(0.003)	(0.061)
Tract housing characteristics
Total housing units (in ‘000s)	0.020	0.344***	0.161*	−0.866
	(0.051)	(0.088)	(0.069)	(1.050)
Tract area (sq. mi)	0.503*	−0.762***	−0.243	−3.042
	(0.224)	(0.189)	(0.204)	(2.301)
Median housing value (in 10,000$)	0.001	0.004**	0.002	−0.010
	(0.001)	(0.001)	(0.001)	(0.030)
Median rent (in 100$)	−0.012	−0.028*	0.012	0.069
	(0.008)	(0.013)	(0.012)	(0.270)
% renter households	0.002	0.012***	0.007***	−0.003
	(0.001)	(0.002)	(0.002)	(0.046)
Median year housing built	0.0004	−0.010***	−0.008***	0.049
	(0.001)	(0.002)	(0.002)	(0.039)
% vacant housing units	0.003	0.009	0.012**	−0.077
	(0.003)	(0.007)	(0.004)	(0.074)
Number of public housing units (in ‘00s)	0.042***	−0.125***	−0.019*	−0.112
	(0.007)	(0.013)	(0.009)	(0.205)
Number of gas shutoff complaints			0.014***
			(0.001)
ρ	0.019***	0.017***	0.010	2.264***
	(0.002)	(0.004)	(0.012)	(0.413)
Constant	1.154	19.918***	14.188***	−122.501
	(2.504)	(4.333)	(3.568)	(76.928)
Observations	2,063	2,105	1,922	1,922

Note: Standard errors in parentheses. All models exclude census tracts with no households and census tracts where the entirety of the population is in group quarters. The analysis of FDNY dispatch data excludes all census tracts in the Highbridge/Concourse and Fordham/University Heights Community Districts of the Bronx as there were no FDNY dispatch data for this area of the city in the NYC Open Data portal. The analysis of gas shutoff complaints only includes non-duplicate complaints for the same incident. The analysis of gas shutoff violations and the amount of time complaints remain open only includes census tracts with at least one gas shutoff complaint.

*** p < 0.001; ** p < 0.01; * p < 0.05 (two-tailed tests).

As the first column of Table 2 shows, the number of reported gas leaks is higher in tracts with higher percent Black, Hispanic, and Asian residents. It is also higher in tracts with more public housing units, in tracts that did not gentrify between 2000 and 2010, and in tracts with greater percent college-educated residents. The number of reported gas leaks is higher in tracts with larger populations and larger areas and lower in tracts with greater percent foreign born residents. The sign of these variables confirms our expectations that the racial and ethnic composition of a tract as well as its socioeconomic make-up is a statistically significant correlate of the number of reported gas leaks. Other variables that prior literature has indicated are related to the “calling propensity” to 311 are either not statistically significant or operate in the opposite direction as expected. For example, the percent of households with children has a negative association with the number of reported gas leaks.

The second column of Table 2 shows that tracts with more Black and Hispanic residents, tracts with more residents in poverty, and tracts that did not gentrify between 2000 and 2010 are more likely to have higher gas shutoff complaints. Given the spatial distribution of gas shutoffs and the high levels of segregation by race and income at the neighborhood level, covariates such as percent foreign born lose statistical significance upon the inclusion of other variables at the tract level. We return to this point in the Sensitivity Analyses section below. On the other hand, tracts with more renters, lower rents, and older housing units are more likely to report gas shutoffs. These predictions are in line with our expectations that neighborhoods with older infrastructure would require more repairs and, thus, would experience higher levels of gas shutoffs. Tracts with more public housing units are less likely to report gas shutoffs. We think that this is an artifact of the 311 system which until 2022 did not allow public housing residents to call in maintenance complaints (Turetsky 2022).

Just as it is the case with gas shutoffs, tracts with more Black and Hispanic residents are more likely to have greater number of shutoff violations issued by the City. Violations are also higher in tracts with higher levels of poverty, tracts that were eligible to gentrify over time, tracts with more renter households, and tracts with an older housing stock. Lastly, we do not find statistically significant relationships between the amount of time that gas shutoff complaints remain open and any of our tract-level variables.

To make the regression results more concrete, we computed the predicted number of gas leaks and shutoffs for a tract where the typical White, Black, and Hispanic person lived in New York City.⁵ The typical White person lived in a tract with about 52 predicted gas leak reports and 26 predicted gas shutoffs (or respectively, about 12 percentage points and 22 percentage points lower than the NYC overall tract average). The typical Black person lived in a tract with 74 predicted gas leak reports and 73 predicted gas shutoffs (or about 1.25 times and 2.21 times the NYC average), while the typical Hispanic person lived in a tract with 65 predicted gas leak reports and 54 predicted gas shutoffs (or 1.1 times and 1.64 times the NYC average). These stark disparities across tracts have important implications for environmental justice in New York City – a point to which we return in the Discussion section below.

In sum, gas leaks and shutoffs in New York City follow longstanding spatial patterns of inequality by race, poverty, and type of housing. Using both visualizations and statistical analysis, we find that it is lower-income communities of color where not only more gas leaks are reported but also where residents are more likely to have their gas shut off. Neighborhoods with large public housing properties are especially likely to report gas leaks. Our analysis demonstrates that the spatial patterning of gas leaks and gas shutoffs has the potential to compound the environmental hazards already present in lower-income communities of color when it comes to both air pollution and inadequate housing.

Sensitivity Analyses

In additional sensitivity analyses we explored the inclusion of other covariates in our models, such as the percent of the tract population receiving public assistance, the percent of the population that does not speak English well or at all, and the percent of the population with professional occupations. In New York City, 311 is the hub for many social safety net programs, so the population receiving public assistance might be more familiar with how to receive services through the system. Residents with professional occupations might be more likely to demand services from the city, while linguistically isolated residents may be less likely to contact 311 due to language barriers. Given the high degree of socioeconomic inequality and residential segregation in NYC, none of these additional variables added to the model's explanatory power, as occupational attainment is highly correlated with the racial and educational composition of tracts, linguistic isolation is highly correlated with percent Hispanic, percent Asian, and percent foreign born, while percent receiving public assistance is highly correlated with tract racial composition and the density of households with children.

We also explored an alternative definition of the tract-level indicators of gentrification, where we computed whether a tract gentrified between 1990 and 2010 rather than between 2000 and 2010. Modeling gentrification over a longer period did not change the sign or statistical significance of other coefficients in the regression. It also did not change the sign or the statistical significance of the indicators for non-gentrifying tracts. It did, however, show that tracts that gentrified between 1990 and 2010 had higher numbers of reported gas leaks compared to those tracts that were not eligible to gentrify. Taken together these results show that regardless of the time horizon for gentrification, it is non-gentrifying tracts that have gas leaks and gas shutoffs that are higher. The relationship between gentrified tracts and gas leaks depends on the period of gentrification. Therefore, we recommend that future analyses pay attention to distinctions between gentrification measured over shorter versus longer periods of time.

We also implemented two sensitivity analyses of the gas shutoff regressions, the results of which we report in Appendix Table 2. In the first sensitivity analysis, we explored the relationship between tract characteristics and gas shutoff complaints that were verified by an inspector of the HPD who managed to contact the person who made the complaint. The results of this regression analysis are virtually the same as the results of our model of all non-duplicate gas shutoff complaints, which gives us confidence that the reported gas shutoffs reflect conditions on the ground. Given the high number of tracts with zero gas shutoff violations, we also explored modeling this outcome using a zero-inflated negative binomial model, which simultaneously fits a logit model of whether a tract had at least one violation and a model of the number of violations in a tract. The estimated coefficients between the negative binomial model presented in Table 2 and the zero-inflated model presented in Appendix Table 2 do not differ in any appreciable way. This suggests that our results are robust to different modeling approaches.

Discussion

Our analysis brings heightened attention to gas leaks as an environmental justice issue from a policy perspective. As Luna and Nicholas (2022) have argued, both the timely repair of gas leaks and the regular public reporting of where they occur is essential for tackling leaks in an equitable manner to address air pollution in low-income communities of color. Therefore, the availability of public data on gas leaks, repair status, emissions, and repair failures can better inform public policy regarding near-term targeted interventions to places where leaks reoccur (Edwards et al. 2021). Massachusetts is the only state where utilities are required to report on gas leaks to the state, but even that database is incomplete (Luna and Nicholas 2022). We are not aware of the existence of parallel reporting mechanisms for gas leaks inside buildings. Therefore, required periodic reports on the location of gas leaks and their repair status can aid in efforts to keep the public safe and reduce methane emissions.

In the absence of such reports, our analysis demonstrates the utility of using 911 and 311 data for the rapid detection of urban infrastructural problems. While there have been several point-in-time surveys of street-level gas leaks in two NYC boroughs (Gallagher et al. 2015; Fischer et al. 2017), we are not aware of efforts to systematically monitor the presence of leaks inside buildings and the timeliness of restoration of gas service. Therefore, the fire department dispatch data and the 311-complaint data could serve as a low-cost way to detect the emergence of new gas leak clusters and monitor the repair of pipes within buildings where the gas has been shut off.

Our findings regarding the spatial patterning of gas leaks and gas shutoffs also have significant implications for the distributive and procedural justice aspects of local environmental policy. We show that from a distributive justice perspective, it is typically in Black and Hispanic tracts, in tracts with large public housing properties, and in tracts that have not gentrified over time where the gas infrastructure is failing. This last result is in line with other research on gentrification which has shown that it is persistent concentrated disadvantage (rather than the influx of higher income and better educated new residents) that is associated with a slew of negative neighborhood conditions, such as higher crime and eviction rates (Desmond and Gershenson 2016). Moreover, the same neighborhoods in our analysis that have the highest rates of gas leaks and gas shutoffs are not only on the State's Disadvantaged Communities (DAC) list, but also have the highest asthma hospitalization rates (Corburn, Osleeb and Porter 2006). In New York City, children who live in public housing have the highest asthma rates out of all children (Northridge et al. 2010). Therefore, the gravity of gas leaks and gas shutoffs for low-income communities of color – and especially public housing complexes – underscores the need for prioritizing disadvantaged communities in infrastructure improvements.

From a procedural justice perspective, our analysis demonstrates some of the unintended consequences of NYC's approach to addressing gas leaks. While the City and local utilities are right to prioritize resident safety especially in the aftermath of two deadly explosions, the patterning of gas shutoffs reveals that the gas infrastructure is most frequently failing in segregated neighborhoods that already suffer from the deleterious effects of decades of environmental harm and disinvestment in infrastructure (Park and Quercia 2020; Taylor 2014). Therefore, the consequences of stepped-up scrutiny of the gas infrastructure both under NYC's streets and inside NYC's buildings fall on low-income communities of color.

On the other hand, we do find that city inspectors issue more violations in disadvantaged neighborhoods and that there are no differences across more disadvantaged and less disadvantaged tracts in the average time that gas shut-off complaints remain open. Without a more direct analysis of the behavior of building inspectors it is difficult to speculate about the underlying process of issuing violations for gas shutoff complaints. Previous research, however, has shown that inspectors tend to issue more violations to professional landlords and see their job as “going after the bad guys” (Bartram 2019). From a procedural justice standpoint, NYC building inspectors may work to redress longstanding inequalities in the housing market by penalizing landlords who do not restore the gas service in their properties in lower-income Black and Hispanic communities. Therefore, unlike the geographic patterning of gas leaks and gas shutoffs, the implementation of city inspections for gas-shutoff violations does not seem to follow discriminatory patterns.

From a procedural justice perspective, it is also important to consider who is more likely to call either 911 or 311 to report a gas leak or gas service shutoff. Our regression analyses indicate that it is lower income communities with older housing where more calls are made. This is in contrast to the typical pattern of calling 311 where it is more advantaged communities that tend to use the 311 service more (Legewie and Schaeffer 2016; Minkoff 2016). As Kontokosta and Hong (2021) have argued, our findings may be due to the fact that we examine calls for a potentially dangerous condition as opposed to a “nuisance” or quality-of-life issue. Therefore, we think it is unlikely that our results are driven by neighborhood “contacting propensity” rather than the actual incidence of gas leaks and gas shutoffs in more disadvantaged areas in NYC.

We do, however, uncover one concerning pattern with respect to gas leak reports. Tracts with higher percent of foreign-born residents are less likely to report gas leaks. Previous research has shown that the U.S. foreign born population is not only less likely to contact the government for services, but also that first-generation immigrants are less likely to call 911 as the U.S. immigration system has become more punitive over time and any potential contact with the police might be perceived as unsafe, especially for those with undocumented family members (Asad and Rosen 2019; Levine and Gershenson 2014). Therefore, both the city and local utilities may need to prioritize gas safety in communities with high number of foreign-born residents where gas leaks might be underreported.

Finally, New York's ambitious climate and environmental justice agenda is expected to face many obstacles when it comes to handling the State's extensive gas infrastructure. The State's own plan to achieve the targets of its 2019 Climate Law calls for the speedy transition of residential gas customers to renewable electricity for heating, hot water, and cooking (New York State Climate Action Council 2022). At the same time, in July 2023, the NYS Public Service Commission (PSC) approved another steep hike in electricity and gas bills over the next three years, some of which will go to replace leak-prone gas pipes (Kinniburgh 2023). Every gas line replacement builds new fossil-fuel infrastructure whose long-term costs can be legally collected from utility customers over the material life of new pipes – typically a period of 50 years (Synapse Energy Economics 2023). As new buildings turn fully electric and as higher-income homeowners make energy upgrades to existing housing, gas costs can dramatically increase for everyone else who cannot afford to electrify their home and is stuck paying for the gas infrastructure others no longer use (Nadel 2023).

Disparities in the geographic patterning of gas leaks and gas outages coupled with the high costs of maintaining the current gas infrastructure suggest that the transition away from gas is not only an environmental justice issue but also an energy justice and housing justice issue. Low-income households and households of color already struggle with utility debt (Hernández 2013). In NYC, the prevalence of housing deficiencies, such as mold, vermin, or peeling paint, is much higher amongst Black and Hispanic households and households making less than $25,000 per year. In fact, more than a third of all rent regulated and public housing units - the housing most prevalent in working-class and immigrant communities - have multiple maintenance deficiencies. These units disproportionately house older adults, children, and people with disabilities (NYC Department of Housing Preservation and Development 2022). As gas shutoffs due to leaks add to the economic burdens of disadvantaged communities and to the cumulative housing deficiencies in the units where they live, prioritizing these neighborhoods in the transition away from fossil fuels can lead to less polluted environments, more energy efficient homes, and lower energy costs for those who currently can least afford to decarbonize their homes but experience the brunt of NYC's failing gas infrastructure.

Supplemental Material

sj-docx-1-uar-10.1177_10780874241235641 - Supplemental material for Gas Leaks, Gas Shutoffs, and Environmental Justice in New York City

Supplemental material, sj-docx-1-uar-10.1177_10780874241235641 for Gas Leaks, Gas Shutoffs, and Environmental Justice in New York City by Yana Kucheva and Ronak Etemadpour in Urban Affairs Review

Footnotes

Acknowledgments

The authors would like to express their gratitude to Dr. Michael Grossberg for his valuable advice on the FDNY data collection and for being a Co-PI on one of the sources of funding for the project. The authors also extend their thanks to MD Rahman, Itay Rubin, and Marina Orzechowski for excellent research assistance. The content of the manuscript is the sole responsibility of the authors. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was partially supported by a grant to Dr. Ronak Etemadpour and Dr. Michael Grossberg from ConEdison under the Gas Safety Training BPL Program (Grant #7W014-04 01) and by a grant to Dr. Yana Kucheva from The City College of New York College-wide Research Vision (CRV) Initiative.

ORCID iD

Yana Kucheva

Supplemental Material

Supplemental material for this article is available online.

Notes

Author Biographies

Yana Kucheva is an associate professor at the Department of Sociology in The City College of New York. Her research focuses on housing policy, residential segregation, and environmental justice.

Ronak Etemadpour's academic tenure includes roles as an assistant professor at the City University of New York - City College and Oklahoma State University. She is currently a Senior Project Manager at Verus Research, where she bridges academic theory and industry practice.

References

Agonafir

Candace

Lakhankar

Tarendra

Khanbilvardi

Reza

Krakauer

Nir

Radell

Dave

Devineni

Naresh

. 2022. “A Machine Learning Approach to Evaluate the Spatial Variability of New York City’s 311 Street Flooding Complaints.” Computers, Environment and Urban Systems 97 (October): 101854. https://doi.org/10.1016/j.compenvurbsys.2022.101854.

Alesina

Alberto

Baqir

Reza

Easterly

William

. 1999. “Public Goods and Ethnic Divisions.” The Quarterly Journal of Economics 114 (4): 1243–84. https://doi.org/10.1162/003355399556269.

Allen

Simon K.

Plattner

Gian-Kasper

Nauels

Alexander

Xia

Stocker

Thomas F.

. 2014. “Climate Change 2013: The Physical Science Basis. An Overview of the Working Group 1 Contribution to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC),” May, 3544.

Anderson

H. Ross

Atkinson

Richard W.

Peacock

Janet

Marston

Louise

Konstantinou

Kostas

. 2004. “Meta-Analysis of Time-Series Studies and Panel Studies of Particulate Matter (PM) and Ozone (O3): Report of a WHO Task Group (No. EUR/04/5046026).” Copenhagen: WHO Regional Office for Europe.

Asad

Asad L.

Rosen

Eva

. 2019. “Hiding Within Racial Hierarchies: How Undocumented Immigrants Make Residential Decisions in an American City.” Journal of Ethnic and Migration Studies 45 (11): 1857–82. https://doi.org/10.1080/1369183X.2018.1532787.

Banzhaf

Spencer

Lala

Timmins

Christopher

. 2019. “Environmental Justice: The Economics of Race, Place, and Pollution.” Journal of Economic Perspectives 33 (1): 185–208. https://doi.org/10.1257/jep.33.1.185.

Bartram

Robin.

2019. “Going Easy and Going After: Building Inspections and the Selective Allocation of Code Violations.” City & Community 18 (2): 594–617. https://doi.org/10.1111/cico.12392.

Brand

David.

2021. “Formerly Homeless Families Go Two Months Without Gas at Harlem Apartment Complex.” CityLimits, July 12, 2021. https://citylimits.org/2021/07/12/formerly-homeless-families-go-two-months-without-gas-at-harlem-apartment-complex/.

Bullard

Robert.

1990. Dumping in Dixie: Race, Class, and Environmental Quality. Boulder, CO: Westview Press.

10.

Burke

Kerry

Gartland

Michael

. 2019. “Upper West Side Tenants Go Months without Gas, Problems with Lights.” NY Daily News, June 23, 2019. https://www.nydailynews.com/new-york/ny-rent-strike-925-columbus-tents-light-gas-20190623-oj4ntiwohvfunjxjeaij7za6ty-story.html.

11.

Cavallo

Sara

Lynch

Joann

Scull

Peter

. 2014. “The Digital Divide in Citizen-Initiated Government Contacts: A GIS Approach.” Journal of Urban Technology 21 (4): 77–93. https://doi.org/10.1080/10630732.2014.942167.

12.

Center for an Urban Future. 2014. “Caution Ahead: Overdue Investments for New York’s Aging Infrastructure.” https://nycfuture.org/pdf/Caution-Ahead.pdf.

13.

Consolidated Edison Company of New York. 2022. “A Comprehensive View of Our Gas System through 2050.” https://documents.dps.ny.gov/public/Common/ViewDoc.aspx?DocRefId=%7B958536EC-3DC6-4C57-93D5-74816B9E4CD7%7D.

14.

Corburn

Jason

Osleeb

Jeffrey

Porter

Michael

. 2006. “Urban Asthma and the Neighbourhood Environment in New York City.” Health & Place 12 (2): 167–79. https://doi.org/10.1016/j.healthplace.2004.11.002.

15.

Desmond

Matthew

Gershenson

Carl

. 2016. “Housing and Employment Insecurity among the Working Poor.” Social Problems 63 (1): 46–67. https://doi.org/10.1093/socpro/spv025.

16.

Di Gregorio

Enrico

Jordan

Felipe

. 2019. “The Diversity We Breath: Community Diversity and Gas Leak Management.” Unpublished manuscript. https://scholar.harvard.edu/files/digregorio/files/the_diversity_we_breath_-_di_gregorio_and_jordan_-_heep.pdf.

17.

Edwards

Morgan R.

Giang

Amanda

Macey

Gregg P.

Magavi

Zeyneb

Nicholas

Dominic

Ackley

Robert

Schulman

Audrey

. 2021. “Repair Failures Call for New Policies to Tackle Leaky Natural Gas Distribution Systems.” Environmental Science & Technology 55 (10): 6561–70. https://doi.org/10.1021/acs.est.0c07531.

18.

Elliott

James R.

Brown

Phylicia Lee

Loughran

Kevin

. 2020. “Racial Inequities in the Federal Buyout of Flood-Prone Homes: A Nationwide Assessment of Environmental Adaptation.” Socius: Sociological Research for a Dynamic World 6 (January): 237802312090543. https://doi.org/10.1177/2378023120905439.

19.

Fischer

Joseph C. von

Cooley

Daniel

Chamberlain

Sam

Gaylord

Adam

Griebenow

Claire J.

Hamburg

Steven P.

Salo

Jessica

Schumacher

Russ

Theobald

David

Ham

Jay

. 2017. “Rapid, Vehicle-Based Identification of Location and Magnitude of Urban Natural Gas Pipeline Leaks.” Environmental Science & Technology 51 (7): 4091–99. https://doi.org/10.1021/acs.est.6b06095.

20.

Freeman

Lance.

2005. “Displacement or Succession?: Residential Mobility in Gentrifying Neighborhoods.” Urban Affairs Review 40 (4): 463–91. https://doi.org/10.1177/1078087404273341.

21.

Gallagher

Morgan E.

Down

Adrian

Ackley

Robert C.

Zhao

Kaiguang

Phillips

Nathan

Jackson

Robert B.

. 2015. “Natural Gas Pipeline Replacement Programs Reduce Methane Leaks and Improve Consumer Safety.” Environmental Science & Technology Letters 2 (10): 286–91. https://doi.org/10.1021/acs.estlett.5b00213.

22.

Giambusso

David.

2014. “Council Unaware FDNY Responds to All Gas-Leak Reports.” Politico, June 18, 2014. https://www.politico.com/states/new-york/city-hall/story/2014/06/council-unaware-fdny-responds-to-all-gas-leak-reports-013781.

23.

Hammel

Daniel J.

Wyly

Elvin K.

. 1996. “A Model for Identifying Gentrified Areas with Census Data.” Urban Geography 17 (3): 248–68. https://doi.org/10.2747/0272-3638.17.3.248.

24.

Henchen

Mike

Kroh

Kiley

. 2020. “A New Approach to America’s Rapidly Aging Gas Infrastructure.” RMI. https://rmi.org/a-new-approach-to-americas-rapidly-aging-gas-infrastructure/#:∼:text=Instead%20of%20continuing%20to%20invest,%2C%20water%20heating%2C%20and%20cooking.

25.

Hernández

Diana.

2013. “Energy Insecurity: A Framework for Understanding Energy, the Built Environment, and Health Among Vulnerable Populations in the Context of Climate Change.” American Journal of Public Health 103 (4): e32–34. https://doi.org/10.2105/AJPH.2012.301179.

26.

Howarth

Robert W.

2014. “A Bridge to Nowhere: Methane Emissions and the Greenhouse Gas Footprint of Natural Gas.” Energy Science & Engineering 2 (2): 47–60. https://doi.org/10.1002/ese3.35.

27.

Howell

Junia

Elliott

James R.

. 2018. “As Disaster Costs Rise, So Does Inequality.” Socius: Sociological Research for a Dynamic World 4 (January): 237802311881679. https://doi.org/10.1177/2378023118816795.

28.

Jackson

Robert B.

Down

Adrian

Phillips

Nathan G.

Ackley

Robert C.

Cook

Charles W.

Plata

Desiree L.

Zhao

Kaiguang

. 2014. “Natural Gas Pipeline Leaks Across Washington, DC.” Environmental Science & Technology 48 (3): 2051–58. https://doi.org/10.1021/es404474x.

29.

Johnson

Steven.

2010. “What a Hundred Million Calls to 311 Reveal About New York.” Wired, 2010. https://www.wired.com/2010/11/ff_311_new_york/.

30.

Kinniburgh

Colin.

2023. “Why Your Energy Bills Are Going Up. New York’s Labyrinthine ‘Rate Case’ Process, Explained.” New York Focus, August 7, 2023. https://nysfocus.com/2023/08/07/energy-bill-rate-hikes-psc-coned.

31.

Konisky

David M.

Reenock

Christopher

. 2018. “Regulatory Enforcement, Riskscapes, and Environmental Justice.” Policy Studies Journal 46 (1): 7–36. https://doi.org/10.1111/psj.12203.

32.

Konisky

David M.

Schario

Tyler S.

. 2010. “Examining Environmental Justice in Facility-Level Regulatory Enforcement: Environmental Justice in Facility-Level Regulatory Enforcement.” Social Science Quarterly 91 (3): 835–55. https://doi.org/10.1111/j.1540-6237.2010.00722.x.

33.

Kontokosta

Constantine E.

Hong

Boyeong

. 2021. “Bias in Smart City Governance: How Socio-Spatial Disparities in 311 Complaint Behavior Impact the Fairness of Data-Driven Decisions.” Sustainable Cities and Society 64 (January): 102503. https://doi.org/10.1016/j.scs.2020.102503.

34.

Ladage

Stefan

Blumenberg

Martin

Franke

Dieter

Bahr

Andreas

Lutz

Rüdiger

Schmidt

Sandro

. 2021. “On the Climate Benefit of a Coal-to-Gas Shift in Germany’s Electric Power Sector.” Scientific Reports 11 (1): 11453. https://doi.org/10.1038/s41598-021-90839-7.

35.

Legewie

Joscha

Schaeffer

Merlin

. 2016. “Contested Boundaries: Explaining Where Ethnoracial Diversity Provokes Neighborhood Conflict.” American Journal of Sociology 122 (1): 125–61. https://doi.org/10.1086/686942.

36.

Lerman

Amy E.

Weaver

Vesla

. 2014. “Staying out of Sight? Concentrated Policing and Local Political Action.” The ANNALS of the American Academy of Political and Social Science 651 (1): 202–19. https://doi.org/10.1177/0002716213503085.

37.

Levine

Jeremy R.

Gershenson

Carl

. 2014. “From Political to Material Inequality: Race, Immigration, and Requests for Public Goods.” Sociological Forum 29 (3): 607–27. https://doi.org/10.1111/socf.12106.

38.

Liang

Jiaqi.

2016. “Defining Environmental Justice Communities for Regulatory Enforcement: Implications from a Block-Group-Level Analysis of New York State: Environmental Justice Areas and Regulatory Enforcement.” Review of Policy Research 33 (6): 666–85. https://doi.org/10.1111/ropr.12202.

39.

Liu

Yan

Cheshire

Lynda

Wang

Siqin

Xuanming

. 2019. “A Socio-Spatial Analysis of Neighbour Complaints Using Large-Scale Administrative Data: The Case in Brisbane, Australia.” Cities 90 (July): 168–80. https://doi.org/10.1016/j.cities.2019.02.010.

40.

Logan

John

Molotch

Harvey

. 1987. Urban Fortunes: The Political Economy of Place. Berkeley, CA: University of California Press.

41.

Long

J. Scott

Freeze

Jeremy

. 2005. Regression Models for Categorical and Limited Dependent Variables. College Station, TX: Stata Press.

42.

Luna

Marcos

Nicholas

Dominic

. 2022. “An Environmental Justice Analysis of Distribution-Level Natural Gas Leaks in Massachusetts, USA.” Energy Policy 162 (March): 112778. https://doi.org/10.1016/j.enpol.2022.112778.

43.

Lala.

2020. “Mapping the Clean Air Haves and Have-Nots.” Science 369 (6503): 503–4. https://doi.org/10.1126/science.abb0943.

44.

McDonald

Yolanda J.

Jones

Nicole E.

. 2018. “Drinking Water Violations and Environmental Justice in the United States, 2011–2015.” American Journal of Public Health 108 (10): 1401–7. https://doi.org/10.2105/AJPH.2018.304621.

45.

Miller

Myles.

2016. “Red Hook NYCHA Residents Prepare Holiday Meals on Hot Plates after Gas Shut Off.” Pix 11 News, December 24, 2016. https://pix11.com/news/local-news/brooklyn/red-hook-nycha-residents-prepare-holiday-meals-on-hot-plates-after-gas-shut-off/.

46.

Minkoff

Scott L.

2016. “NYC 311: A Tract-Level Analysis of Citizen–Government Contacting in New York City.” Urban Affairs Review 52 (2): 211–46. https://doi.org/10.1177/1078087415577796.

47.

Mohai

Paul

Pellow

David

Roberts

J. Timmons

. 2009. “Environmental Justice.” Annual Review of Environment and Resources 34 (1): 405–30. https://doi.org/10.1146/annurev-environ-082508-094348.

48.

Morris

Jackson.

2019. “Unpacking New York’s Big New Climate Bill: A Primer.” Natural Resources Defence Coalition.

49.

Nadel

Steven.

2023. “Impact of Electrification and Decarbonization on Gas Distribution Costs.” American Council for an Energy-Efficient Economy. https://www.aceee.org/sites/default/files/pdfs/U2302.pdf.

50.

New York State Climate Action Council. 2022. “New York State Climate Action Council Scoping Plan.” climate.ny.gov/ScopingPlan.

51.

New York State Research and Development Authority. 2022. “New York State Climate Action Council Finalizes Scoping Plan to Advance Nation-Leading Climate Law.” https://www.nyserda.ny.gov/About/Newsroom/2022-Announcements/2022-12-19-NYS-Climate-Action-Council-Finalizes-Scoping-Plan-to-Advance-Nation-Leading-Climate-Law#:∼:text=The%20Plan%20outlines%20actions%20needed,zero%20emissions%20statewide%20by%202050.

52.

New York State Research and Development Authority. 2023. “Disadvantaged Communities Criteria.” https://climate.ny.gov/resources/disadvantaged-communities-criteria/.

53.

Nixon

Rob.

2013. Slow Violence and the Environmentalism of the Poor. Cambridge, MA: Harvard University Press.

54.

Northridge

Jennifer

Ramirez

Olivia F.

Stingone

Jeanette A.

Claudio

Luz

. 2010. “The Role of Housing Type and Housing Quality in Urban Children with Asthma.” Journal of Urban Health 87 (2): 211–24. https://doi.org/10.1007/s11524-009-9404-1.

55.

NYC Department of Housing Preservation and Development. 2022. “2021 New York City Housing and Vacancy Survey Selected Initial Findings.” https://www.nyc.gov/site/hpd/about/research.page.

56.

NYC Office of Technology & Innovation. 2022. “Customer Satisfaction Surveys in Designated Citywide Languages: 2021 Final Results.” https://www1.nyc.gov/assets/oti/downloads/pdf/reports/311-customer-satisfaction-by-language-2021.pdf.

57.

NYC Office of the Mayor. 2016. “Mayor de Blasio Signs Package of Legislation to Increase Gas Safety in NYC Buildings.” December 6, 2016. https://www1.nyc.gov/office-of-the-mayor/news/929-16/mayor-de-blasio-signs-package-legislation-increase-gas-safety-nyc-buildings.

58.

NYU Furman Center. 2016. “State of New York City’s Housing and Neighborhoods in 2015.” https://furmancenter.org/research/sonychan/2015-report.

59.

Oldershausen

Sasha von

. 2020. “Surviving on Hot Plates and Takeout: 73 Days Without Cooking Gas.” The New York Times, December 4, 2020, sec. MB.

60.

Park

Kevin A.

Quercia

Roberto G.

. 2020. “Who Lends Beyond the Red Line? The Community Reinvestment Act and the Legacy of Redlining.” Housing Policy Debate 30 (1): 4–26. https://doi.org/10.1080/10511482.2019.1665839.

61.

Pevzner

Nicholas.

2018. “Buried Grudges.” Urban Omnibus, October 11, 2018. https://urbanomnibus.net/2018/10/buried-grudges/.

62.

Phillips

Nathan G.

Ackley

Robert

Crosson

Eric R.

Down

Adrian

Hutyra

Lucy R.

Brondfield

Max

Karr

Jonathan D.

Zhao

Kaiguang

Jackson

Robert B.

. 2013. “Mapping Urban Pipeline Leaks: Methane Leaks Across Boston.” Environmental Pollution 173 (February): 1–4. https://doi.org/10.1016/j.envpol.2012.11.003.

63.

PHMSA. 2022. “Cast and Wrought Iron Inventory.” 2022. https://www.phmsa.dot.gov/data-and-statistics/pipeline-replacement/cast-and-wrought-iron-inventory.

64.

Plant

Genevieve

Kort

Eric A.

Floerchinger

Cody

Gvakharia

Alexander

Vimont

Isaac

Sweeney

Colm

. 2019. “Large Fugitive Methane Emissions From Urban Centers Along the U.S. East Coast.” Geophysical Research Letters 46 (14): 8500–7. https://doi.org/10.1029/2019GL082635.

65.

Riedel

Natalie

Scheiner

Joachim

Müller

Grit

Köckler

Heike

. 2014. “Assessing the Relationship Between Objective and Subjective Indicators of Residential Exposure to Road Traffic Noise in the Context of Environmental Justice.” Journal of Environmental Planning and Management 57 (9): 1398–421. https://doi.org/10.1080/09640568.2013.808610.

66.

Sampson

Robert J.

Morenoff

Jeffrey D.

Gannon-Rowley

Thomas

. 2002. “Assessing ‘Neighborhood Effects’: Social Processes and New Directions in Research.” Annual Review of Sociology 28 (1): 443–78. https://doi.org/10.1146/annurev.soc.28.110601.141114.

67.

Sampson

Robert J.

Raudenbush

Stephen W.

Earls

Felton

. 1997. “Neighborhoods and Violent Crime: A Multilevel Study of Collective Efficacy.” Science 277 (5328): 918–24. https://doi.org/10.1126/science.277.5328.918.

68.

Sánchez

Cecilia A

Rios

María Jazmín

Murray

Maureen H

. 2021. “Social and Environmental Correlates of Rat Complaints in Chicago.” Journal of Urban Ecology 7 (1): juab006. https://doi.org/10.1093/jue/juab006.

69.

Santora

Marc.

2014. “At Least 3 Killed as Gas Explosion Hits East Harlem.” The New York Times, March 12, 2014. https://www.nytimes.com/2014/03/13/nyregion/east-harlem-building-collapse.html.

70.

Schaider

Laurel A.

Swetschinski

Lucien

Campbell

Christopher

Rudel

Ruthann A.

. 2019. “Environmental Justice and Drinking Water Quality: Are There Socioeconomic Disparities in Nitrate Levels in U.S. Drinking Water?” Environmental Health 18 (1): 3. https://doi.org/10.1186/s12940-018-0442-6.

71.

Schwarz

Kirsten

Fragkias

Michail

Boone

Christopher G.

Zhou

Weiqi

McHale

Melissa

Grove

J. Morgan

O’Neil-Dunne

Jarlath

, et al. 2015. “Trees Grow on Money: Urban Tree Canopy Cover and Environmental Justice.” Edited by Steven Arthur Loiselle. PLOS ONE 10 (4): e0122051. https://doi.org/10.1371/journal.pone.0122051.

72.

Scott

Ryan P.

Scott

Tyler A.

Greer

Robert A.

. 2019. “The Environmental and Safety Performance of Gas Utilities in the United States.” Energy Policy 133 (October): 110892. https://doi.org/10.1016/j.enpol.2019.110892.

73.

Small

Eddie.

2020. “No Gas in Your Apartment? Don’t Hold Your Breath.” The Real Deal, January 13, 2020. https://therealdeal.com/2020/01/13/no-gas-in-your-apartment-dont-hold-your-breath/.

74.

Solis

Gustavo.

2015. “In Wake of Gas Explosion, Con Ed Spending $1B to Replace Leak-Prone Pipes.” DNA Info, December 4, 2015. https://www.dnainfo.com/new-york/20151204/east-harlem/wake-of-gas-explosion-con-ed-spending-1b-replace-leak-prone-pipes/.

75.

Stempel

Jonathan.

2017. “Con Edison Reaches $153 Million Settlement over Fatal 2014 Harlem Blast.” Reuters, February 16, 2017. https://www.reuters.com/article/us-new-york-conedison/con-edison-reaches-153-million-settlement-over-fatal-2014-harlem-blast-idUSKBN15V2Q8.

76.

Synapse Energy Economics. 2023. “The High Cost of New York Gas Utilities’ Leak-Prone Pipe Replacement Programs.” https://www.synapse-energy.com/sites/default/files/22-017_High_Cost_NY_Gas_Utilities_LPP_Programs_0.pdf.

77.

Taylor

Dorceta.

2014. Toxic Communities: Environmental Racism, Industrial Pollution, and Residential Mobility. New York, NY: NYU Press.

78.

Turetsky

Doug.

2022. “Under New Law, NYCHA Tenants Can Log Maintenance Complaints with 311. But Advocates Want More Substantial Changes.” CityLimits, March 21, 2022. https://citylimits.org/2022/03/21/under-new-law-nycha-tenants-can-log-maintenance-complaints-with-311-but-advocates-want-more-substantial-changes/.

79.

Vardi

Itai.

2021. “While Touting Clean Energy Steps, Con Edison Continues to Invest in Gas Infrastructure.” Energy and Policy Institute. https://energyandpolicy.org/while-touting-clean-energy-steps-con-edison-continues-to-invest-in-gas-infrastructure/.

80.

Walsh

Michael

Bloomberg

Michael

. 2023. “The Future of Gas in New York State.” Building Decarbonization Coalition.

81.

Ward

Michael

Gleditsch

Kristian

. 2008. Spatial Regression Models. 2455 Teller Road, Thousand Oaks California 91320 United States of America: Sage Publications, Inc. https://doi.org/10.4135/9781412985888.

82.

Weitzman

Michael

Baten

Ahmareen

Rosenthal

David G.

Hoshino

Risa

Tohn

Ellen

Jacobs

David E.

. 2013. “Housing and Child Health.” Current Problems in Pediatric and Adolescent Health Care 43 (8): 187–224. https://doi.org/10.1016/j.cppeds.2013.06.001.

83.

Weller

Zachary D.

Yang

Duck Keun

von Fischer

Joseph C.

. 2019. ““An Open Source Algorithm to Detect Natural Gas Leaks from Mobile Methane Survey Data.” Edited by Matthias Mauder.” PLOS ONE 14 (2): e0212287. https://doi.org/10.1371/journal.pone.0212287.

84.

West

J. Jason

Fiore

Arlene M.

Horowitz

Larry W.

Mauzerall

Denise L.

. 2006. “Global Health Benefits of Mitigating Ozone Pollution with Methane Emission Controls.” Proceedings of the National Academy of Sciences 103 (11): 3988–93. https://doi.org/10.1073/pnas.0600201103.

85.

Whitford

Emma.

2016. “Sunset Park Tenants Are Living Without Gas As Gentrification Closes In.” Gothamist, January 22, 2016. https://gothamist.com/news/sunset-park-tenants-are-living-without-gas-as-gentrification-closes-in.

86.

Willis

Mary D.

Cushing

Lara J.

Buonocore

Jonathan J.

Deziel

Nicole C.

Casey

Joan A.

. 2023. “It’s Electric! An Environmental Equity Perspective on the Lifecycle of Our Energy Sources.” Environmental Epidemiology 7 (2): e246. https://doi.org/10.1097/EE9.0000000000000246.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.99 MB