Research · Household Burden & Lock-In · September 2026

The cost of waiting

Charging policy creates two analytically distinct costs for residents who cannot use residential electricity where they park. First, an EV owner pushed onto public charging pays a cash premium and a time tax, every year. Second — and longer-lived — if that burden tips a household's vehicle-replacement decision toward gasoline, the resulting vehicle stays on the road for most of a decade after the policy finally changes. This page summarizes the paper's findings and lets you test its assumptions in two live models.

The findings in brief

1. Residential electricity is the low-cost EV fuel. The same 3,600 kWh a year costs about $468 at home, $1,080 at the favorable public-L2 price, and $1,440 at DC fast rates — a direct cash premium of roughly $612–$972 per household per year before valuing a minute of anyone's time.
2. The bigger hidden cost is time. Even deliberately favorable public-charging scenarios impose roughly 67–78 nonproductive hours per household per year — worth $1,294–$1,515/yr at USDOT's general personal-travel value, $2,588–$3,030 at USDOT's walking/waiting value, or $3,971–$4,650 at the Alexandria income proxy.
3. The burden scales citywide. Under the current settings in Model 1 below, the aggregate burden is about $2.8M in extra electricity and 352,000 hours per year — hours are a real household burden even though no City budget line records them.
4. Delay locks in gasoline vehicles. In the central scenario, each year of delay is associated with about 215 additional gasoline-vehicle purchases and ~1,804 committed gasoline vehicle-years — roughly $12,570 in foregone operating savings and 21.2 metric tons of operational CO₂ per induced vehicle over its 8.4-year holding period.
5. Cross-curb is one part of a portfolio. Multifamily, workplace, public L2, and DC fast charging remain necessary; this analysis asks only whether households with a feasible safe residential connection should be required to consume shared infrastructure instead.

Model 1 — the charging-access burden

What does it cost Alexandria EV households when they cannot use residential electricity where they park? Fixed assumptions: 12,000 miles/yr, 0.30 kWh/mile (3,600 kWh/yr); $0.13/kWh residential, $0.30/kWh public L2 (a deliberately favorable price — see the break-even analysis), $0.40/kWh DC fast; 78.1 nonproductive hours/yr for mature L2, 66.7 for optimistic DC fast.

Model 2 — vehicle lock-in from policy delay

Vehicle purchases are durable: owners keep a new car an average of 8.4 years. If the charging barrier changes a replacement decision from EV to gasoline, changing the policy a year later does not reverse the purchase — the household carries the operating costs, and the air carries the emissions, for the vehicle's whole holding period. Committed vehicle-years measures that full future exposure from purchases induced during the delay; realized by enactment counts only the burden that has already accumulated by the day the policy changes. A policy change stops new lock-in; it does not undo the old.

What the buildout itself would cost

The household model and the infrastructure-capital model answer different questions, and adding charger capital to the household price would double-count it — a viable retail charging rate already embeds capital recovery (see the break-even analysis). But the physical capital is worth reporting on its own, because it shows the scale of infrastructure that must be financed before those retail rates can exist at all. Using the capacity model's 25%-penetration case (about 1,502 public L2 ports or 160 DC fast ports for the shared-charging-dependent cohort):

System capital requirements, not City expenditures. Actual site costs vary materially with utility interconnection, make-ready, trenching, permitting, and transformer work. The $150,000-per-plug DC fast figure is NREL's order-of-magnitude planning benchmark.

Home charging is not free either, and a fair comparison says so: the residential model assumes a $1,500 installed charger and a $300 removable cable ramp — $1,800 of household capital before any insurance requirement, a private asset rather than shared infrastructure. Where a household prefers a permanent flush crossing instead of a ramp, the companion crossing-options analysis uses a provisional Alexandria planning range of about $1,250–$2,250 for a recessed cable channel (central ~$1,750, excluding the EVSE) — a planning estimate, not a contractor quote. Cross-curb access therefore shifts capital from shared public infrastructure to household assets that participating residents finance themselves, while roughly halving the shared build-out.

What the models deliberately do not claim

• The 15% addressable share is provisional scenario analysis; a measured Alexandria estimate requires parcel, curb-frontage, parking, and vehicle-registration GIS data.
• The 5/10/15-point adoption effects are sensitivities anchored in the literature (a 2026 Transportation Research Part D household study finds home-charging access instrumental to adoption; a 2026 Scottish causal analysis — a preprint — finds a 2.3-point absolute effect while warning that naive models overstate it), not Alexandria forecasts.
• The CO₂ figures are operational, not lifecycle — though DOE's GREET model finds a representative 2025 EV has ~46% lower lifecycle emissions than a comparable gasoline car, so the direction survives a full accounting.
• Public-health effects of removing tailpipe emissions are real but geographically specific; the paper reports them qualitatively rather than monetizing them.
• The public-charging cases are deliberately favorable scenarios, not descriptions of any particular user's behavior — real households will mix home, workplace, destination, and fast charging.

Sources