Mamdani’s Bill Could Cost $5 Per AMAZON Order

Zohran Mamdani smiling indoors near a voting sign with an American flag
Photo: Ron Adar / Shutterstock

The fight over New York City’s Delivery Protection Act is not really about a $5 surcharge; it is about who owns the last mile — the costs, the risks, and the accountability — in a system built to make “free” delivery feel inevitable.

At a Glance

  • Amazon warns the bill could add up to $5 per NYC order and disrupt its contractor-based delivery network.
  • The legislation would license last‑mile facilities and require direct employment of delivery drivers, curbing subcontracting.
  • Supporters argue the current model shifts safety, wage, and liability risks onto subcontractors and workers.
  • The $5 figure is Amazon’s projection; an independent cost model and city fiscal analysis are still absent from public view.

What the bill targets: the operating core of “free” delivery

For nearly a decade, Amazon has grown a Delivery Service Partner (DSP) network: legally independent small carriers that run branded vans on Amazon routes, under Amazon software, metrics, and uniforms. The model’s appeal is strategic — flexibility, variable capacity, and insulation from direct employment obligations — while still exerting granular operational control day to day. Intro 518 goes straight at that architecture. It would require city licensing for last‑mile facilities and, most consequentially, direct employment of drivers making deliveries in New York City, displacing the subcontractor structure that Amazon has relied on since 2018. In plain terms, the bill tries to match control with responsibility, by aligning the employer of record with the entity setting quotas, routes, and standards.

Amazon’s response has been unusually specific, by corporate standards. The company has told New York customers that the legislation could add up to $5 per order, slow deliveries, and jeopardize its use of more than 40 local delivery businesses employing over 5,000 New Yorkers. It has also warned that it would consider moving certain operations outside the five boroughs, with knock‑on effects for speed and cost. The mechanism behind these claims is not mysterious: direct employment typically raises per‑route labor cost, changes scheduling dynamics, and imposes benefit, retention, and safety compliance that sit today on contractors’ books. The company is saying those costs do not disappear; they reattach to the brand promising one‑day shipping.

How the cost claim stacks up

Two things can be true at once: the bill would increase Amazon’s direct labor and compliance costs in the city, and the headline “up to $5 per order” remains a scenario, not a verified estimate. Amazon has not published the underlying route economics, wage and benefit assumptions, or sensitivity testing behind that number; media reports confirm the figure is the company’s projection rather than an independently audited model. The scale of potential impact is, however, operationally plausible given the shift from a contractor rate card to an employer wage‑benefit structure across thousands of routes — but plausibility is not proof, and the absence of a public model makes it impossible for outsiders to judge pass‑through rates to customers versus margin absorption or service redesign.

The most concrete data point Amazon has offered is exposure: more than 40 DSPs and over 5,000 workers tied to its NYC last‑mile operations. Those counts frame the order‑of‑magnitude, even if they do not dictate the outcome. Some portion of those drivers could, in theory, transition to Amazon payroll; others could remain with contractors serving non‑covered routes; still others could exit if DSPs lose viability. Without a facility‑level compliance plan and workforce transition map, the jobs effect should be read as risk range, not a foregone conclusion.

What supporters want to fix: control without accountability

Backers of the bill argue the current contractor regime allows large retailers to script the work — uniforms, routes, quotas, hours — while disclaiming responsibility when things go wrong. They contend that licensing, safety and training standards, and direct employment would improve pay, benefits, and accountability for injuries and accidents, bringing Amazon under the same obligations as firms that directly employ their drivers. Worker testimony and city messaging underscore the core claim: Amazon branding and discipline operate as if the drivers were Amazon’s, but liability often stops at the subcontractor’s door. In public videos, drivers describe mandatory Amazon‑branded uniforms and discipline for non‑compliance, which supporters cite as evidence of de facto control inside a legal structure that avoids direct employment.

The accountability case is strengthened by a straightforward governance logic: if a company sets production metrics that influence driving behavior and vehicle utilization, the company should bear the compliance and safety duties that follow from those choices. That logic, not an enumerated safety statistic, is the center of gravity for proponents. The bill’s licensing framework through the Department of Consumer and Worker Protection is designed to translate that logic into oversight, with enforceable standards rather than aspirational codes.

Where the real disagreement lies: economics versus incidence

Strip away the rhetoric and the dispute reduces to incidence — who pays and how much. Direct employment raises the transparency of labor costs and likely raises the average cost per completed stop inside city limits. The open questions are distributional. Do those costs land on customers as a surcharge, on the company as lower margins, on service levels through slower delivery or higher minimum‑basket thresholds, or on the network architecture via suburban staging and longer in‑city stems? Amazon says the cost would materialize for customers, up to $5 per order; it has not shown how much would be offset by redesign or absorbed on the P&L. Supporters, for their part, have not produced a counter‑model quantifying safety and retention benefits relative to unit cost, nor a demonstration that direct employment would be cost‑neutral at city scale. The result is a sharp policy trade: certain accountability now versus uncertain price impacts later.

This is familiar political economy. Last‑mile networks chase density and time windows; any policy that raises the marginal cost of a stop invites operational substitution — later windows, consolidated drops, micro‑hubs just beyond the jurisdictional line, or minimum‑order thresholds. Amazon has flagged relocation of operations outside the city as a consideration if the bill passes; this is a consequential but conditional claim, unsupported by a facility map or timeline, and should be treated as leverage until paired with a concrete plan.

How to evaluate the “$5” without the spreadsheet

In the absence of a disclosed model, look for three signals. First, elasticity: If delivery demand is inelastic at small surcharges, pass‑through is likelier; if order frequency is price‑sensitive, Amazon has incentives to bury cost in margins or reconfigure service. Second, redesign options: The more a firm can consolidate stops, shift cut‑off times, or pre‑position volume at the boundary, the more it can blunt a per‑order increase. Third, labor transition: If existing DSP drivers transition into Amazon employment at scale, training and safety costs rise up front but may yield lower injury and turnover costs later — savings that do not show in first‑year arithmetic. None of these factors falsifies Amazon’s headline; they simply determine how much of it survives contact with operations.

The prudent stance, then, is disciplined agnosticism on the exact dollar figure paired with confidence about directionality: direct employment and licensing will raise Amazon’s on‑the‑books costs per NYC order; the magnitude and customer exposure require an independent model to adjudicate. Amazon’s projection is a stake in the ground, not the last word.

What it means for New Yorkers

Expect trade‑offs to surface in practical ways you can see: tighter order cut‑offs, more two‑day defaults, or minimum‑order thresholds for no‑rush shipping; alternatively, a visible surcharge on speed tiers with standard delivery kept flat. If the Council enacts Intro 518 substantially as proposed, city agencies will need to publish implementation guidance that clarifies licensing scope, transition timelines for drivers, and enforcement posture. Business groups have already signaled litigation, which could delay or narrow the rule set and prolong uncertainty for both customers and workers. The fastest path to clarity is sunlight: Amazon should release the assumptions behind its per‑order estimate, and the Council should publish a fiscal and operational analysis showing modeled impacts across compliance scenarios. Absent that, the debate will remain a contest of instincts disguised as forecasts.

Sources:

lehighvalleylive.com, aboutamazon.com, fidelity.com, yahoo.com, cnbc.com, geekwire.com, gisweb.vwd.com, kpvi.com