Inequitable Access to EV Charging Infrastructure

• Charging access and affordability are prominent EV adoption barriers, and public charging availability matters to consumers and infrastructure investors.Insufficient infrastructure can suppress EV demand and discourage private investment.Source: §I · p. 1
• Public charging roll-outs should distribute costs and benefits equitably because their burden disproportionately affects low- and middle-income, underrepresented, and disadvantaged communities.The paper frames equitable EV access as an environmental-justice concern.Source: §I · p. 1
• Prior work examines disparities in EV charging access, but much of it offers qualitative discussion or analyzes isolated socioeconomic factors.The cited literature includes findings on race, income, and access near multi-unit housing.Source: §I · p. 1
• The paper addresses the stated gap with a systematic analysis of correlations among demographic features and charging access and affordability at NYC zip-code granularity.The authors describe this granularity as suited to an urban-justice setting.Source: §I · p. 2
• The study examines correlations between charging-station distribution and interdependent demographic and transportation features, not causal effects.Features include income, poverty, racial makeup, highways, and population size.Source: §I · p. 2

• The EV-station dataset counts stations rather than ports, excludes residential charging, and omits stations for other alternative fuels.The dataset is the US Department of Energy Alternative Fuel Station Locator.Source: §II · p. 2
• Demographic measures come from 2019 ACS one-year estimates and include median household income, poverty rate, and racial-group population percentages by zip code.The paper also uses ACS population data.Source: §II · p. 2
• Major-roadway routes are mapped to NYC zip codes using state roadway data and city zip-code boundaries.Fig. 1 maps the five boroughs and commercial and non-commercial traffic routes.Source: §II · p. 2, Figure 1 · p. 2

• The analysis uses two charging outcomes: whether a zip code has at least one station and the total number of stations there.The hypothesis test compares 100 zip codes with stations against 80 without them.Source: §III · p. 2
• Two-sample t-tests compare average feature values between zip codes with and without stations, using p ≤ 0.05 as the significance threshold.The null hypothesis is equal group means.Source: §III · p. 2
• Conditional analysis examines station counts by the population share identifying as White or non-White, conditioned on a $64,000 annual median-income threshold.The threshold is the estimated NYC median income for 2015–2019.Source: §III · pp. 2–3

• The case study combines zip-code demographic and station data, tests differences between zip codes with and without stations, and uses conditional analysis to examine demographic interdependencies.Fig. 3 depicts the zip-code distribution of charging stations; Table I describes correlations between demographic and transportation features and station presence and counts.Source: §IV · p. 3, Figure 3 · p. 3, Table I · p. 3
• Brooklyn has some of NYC's highest population densities but few charging stations in the same areas, while stations concentrate in lower-density Manhattan.The authors conclude that population density is not a good indicator of station distribution.Source: §IV · p. 3

A. Correlation Analysis

• Median household income and White-identifying population have the strongest positive correlations with EV charger presence and counts, with income more strongly correlated.Higher income is associated with a greater probability of at least one station and a higher station count.Source: A. Correlation Analysis · p. 3, Figure 4 · p. 4
• Manhattan zip codes have the highest median income and station counts, while Bronx zip codes have the lowest; income explains the station distribution throughout NYC.The income–station count correlation is 0.58.Source: Figure 4 · p. 4, A. Correlation Analysis · p. 4
• Highway presence has a weak positive correlation of 0.32 with the number of EV charging stations in a zip code.Source: Figure 4 · p. 4
• White-share patterns do not hold in parts of Staten Island and Brooklyn, where high White-identifying shares coincide with low charger penetration.Source: A. Correlation Analysis · p. 4, Figure 4 · p. 4

B. Hypothesis Testing

• Zip codes with at least one station have significantly higher mean household income and White-identifying population shares, and a lower mean Black-identifying population share, than zip codes without stations.Source: B. Hypothesis Testing · p. 4
• The means of income, White- and Black-identifying population shares, and highway presence differ significantly between the two zip-code groups.Source: B. Hypothesis Testing · p. 4
• Highway presence is the most significant feature distinguishing zip codes with and without chargers, followed by median income and White- and Black-identifying population shares.The analysis reports significant t-values and p-values for these features.Source: B. Hypothesis Testing · p. 4

C. Conditional Analysis

• Above $64,000 median income, EV charging station counts are positively correlated with the percentage of White-identifying residents.Source: C. Conditional Analysis · p. 4
• Table II reports hypothesis tests comparing zip codes with and without EV charging stations.Source: Table II · p. 5
• At or below $64,000 median income, EV charging station counts are negatively correlated with the percentage of non-White-identifying residents.Source: Figure 7 · p. 5
• The conditional findings suggest that understanding station distribution requires intersecting economic and racial population profiles.Source: Figure 7 · p. 5

V. CONCLUSION

• In NYC, station availability and affordability correlate with household income, White-identifying population share, and highway presence, but not population density.Source: §V · p. 5
• The current distribution is heavily skewed against low-income, Black-identifying, and disadvantaged neighborhoods.Source: §V · p. 5
• Existing inequities combined with underdeveloped infrastructure in vulnerable communities may result in biased and inequitable future station siting.Source: §V · p. 5
• The authors argue that justice-centric policy frameworks are imperative for large-scale, equitable EV adoption.Source: §V · p. 5