ICLE Issue Brief

The Middleman Tax on Car Buyers: How Dealer Franchise Laws Raise Vehicle Prices

Executive Summary

Many state automobile-dealer franchise laws require manufacturers to sell vehicles through franchised dealers, even when manufacturers would prefer to sell directly to consumers. Originally enacted to regulate bargaining relationships between manufacturers and their dealer networks, these laws now function as a mandatory intermediary requirement. The result is effectively a “middleman tax” on car buyers: a legally imposed distribution layer that increases vehicle prices even where more efficient models are available.

Using an analytical framework first developed in a 2000 Goldman Sachs study and later cited by the U.S. Department of Justice, this issue brief updates the methodology with current vehicle prices, interest rates, inventory data, and dealer operating costs. The results suggest that direct distribution could generate savings of roughly $3,934 to $4,992 per vehicle. Applied to the modern average transaction price of about $50,000, this represents an efficiency gain of 7.87% to 9.87%—the practical equivalent of removing a substantial implicit surcharge from every new-car purchase.

Mandatory dealer intermediation also fragments the national market, burdens interstate commerce, and fits poorly with modern vehicle business models built around digital retailing, centralized pricing, and software-defined vehicles. The policy principle is simple: allow channel competition. States should not mandate a single distribution architecture when multiple models can compete to serve consumers. Removing the “middleman tax” would lower costs, expand consumer choice, and better align the law with the modern automobile market.

I. Introduction

For most Americans, buying a car is one of the largest purchases they will ever make. Yet in all 50 states, a web of dealer-franchise laws dictates how that purchase must occur. Legislatures enacted these statutes between the 1930s and 1970s to protect franchised dealers from coercive manufacturer practices. Today, they serve a different function: many state laws prohibit manufacturers from selling vehicles directly to consumers—even when no franchisee exists and no vertical relationship requires regulation.[1]

The core question is straightforward: why should state law force an out-of-state manufacturer to adopt a particular retail structure as the price of access to a state’s consumers? This issue brief argues that such mandates impose a large and unnecessary “middleman tax” on car buyers, fragment the national market in ways that burden interstate commerce, and fit poorly with modern direct-to-consumer and software-driven vehicle business models. Based on updated estimates, the current burden imposed by state franchise laws amounts to roughly $3,934 to $4,992 per vehicle—between 7.87% and 9.87% of the average $50,000 purchase price.[2]

This issue brief proceeds in three parts. First, it quantifies the economic costs created by this middleman tax. It builds on a conservative baseline established in a 2009 paper by the U.S. Department of Justice’s Economic Analysis Group[3] and updates that methodology using current vehicle prices, inventory data, interest rates, and compliance costs. Second, it examines the constitutional limits that the Dormant Commerce Clause places on state efforts to dictate firm organization as a condition of market access. Third, it shows that many state franchise statutes fit poorly—both textually and structurally—with the business models they now attempt to regulate. Courts and regulators often rely on expansive interpretations of provisions designed for a fundamentally different market.

At bottom, this story is about affordability and consumer welfare. Direct sales are not a luxury preference. They lower acquisition costs, reduce transaction frictions, and speed consumer access to cost-saving vehicle technologies. In a market where the average new-vehicle transaction price now exceeds $50,000[4] and affordability sits near historic lows,[5] the stakes of this debate are substantial.

II. The Origins and Evolution of Dealer-Franchise Laws

Automobile-dealer franchise laws emerged between the 1930s and 1950s, when the U.S. auto market was dominated by the “Big Three”: General Motors, Ford, and Chrysler.[6] In that environment, manufacturers were widely seen as holding overwhelming bargaining power over their franchised dealer networks. Critics argued that automakers could force dealers to accept unwanted vehicles, terminate franchise agreements without cause, or place new franchises inside an existing dealer’s territory. States responded by enacting franchise-protection statutes. By the mid-20th century, every state had adopted some form of dealer-protection law.

These laws targeted a specific problem: the balance of power within an intrabrand relationship. Legislatures sought to regulate the vertical relationship between a manufacturer and its own franchised dealers. The entire statutory framework assumed the existence of that relationship and focused on the bargaining power within it.[7]

The modern automobile market looks very different. The Big Three’s combined market share has fallen from roughly 90% in the 1950s to about 40% today.[8] Interbrand competition now defines the market, with dozens of manufacturers competing for consumers. Dealers have also changed. What were once largely small, family-owned businesses have evolved into large, sophisticated enterprises, often owned by multi-state groups or publicly traded companies. As of 2024, the United States had about 16,957 franchised light-vehicle dealerships.[9] Currently, about 150 dealerships control roughly 30% of industry revenue, and some estimates project that figure will reach 50% by 2050.[10]

This shift highlights a categorical distinction. Regulating an existing vertical relationship between a manufacturer and its franchisees differs fundamentally from prohibiting an alternative distribution model where no franchisee exists. New entrants such as Tesla, Lucid, Rivian, Scout, and others operate without franchised dealers. There is no intrabrand relationship to regulate—a point recognized by at least one state supreme court.[11] Applying franchise-protection statutes to these firms therefore reflects a category error. It conflates the regulation of an existing franchise relationship with the prohibition of a competing business model.

The persistence of these laws, despite the disappearance of the market conditions that produced them, illustrates a classic pattern of regulatory capture.[12] Automobile dealers—the organized interest group that benefits from the statutes—have used the legislative process to preserve economic rents, while spreading the resulting costs across consumers.

III. The Economic Cost of the ‘Middleman Tax’

Dealer-franchise laws operate, in economic terms, as a structural restraint on distribution. They require a two-tier channel—manufacturer to dealer to consumer—even where a one-tier channel—manufacturer to consumer—would be feasible and potentially more efficient. This requirement effectively imposes what this issue brief refers to as a “middleman tax” on vehicle purchases.

The resulting costs arise from several sources. Maintaining dealer inventories ties up capital and slows inventory turnover. Sales commissions and related transaction overhead increase the cost of each sale. Thousands of retail locations duplicate facilities, staffing, and administrative functions that a more centralized distribution system could avoid. Consumers also bear bargaining and search costs that accompany dealership negotiations. Finally, the make-to-stock distribution model weakens the link between production and consumer preferences, producing misallocation costs that appear as rebates, incentives, and markdowns on slow-moving inventory.[13] Although not captured in the foregoing analysis as an economic cost, the dealership model also imposes substantial transaction costs on consumers. The average vehicle purchase required roughly 13 hours of buyer time in 2025, including research, dealer visits, and negotiation.[14]

The sections that follow quantify these effects and place them in a broader economic context. Section III.A summarizes the baseline estimates developed by the U.S. Department of Justice’s Economic Analysis Group. Section III.B updates those estimates using current vehicle prices, interest rates, and dealer operating costs, producing a modern estimate of the middleman tax imposed by dealer mandates. Section III.C then examines the dynamic effects of these rules, showing how dealer mandates can dampen manufacturing investment and constrain innovation in software-defined vehicles and integrated digital-service platforms.

A. The DOJ Baseline Estimate

The most rigorous government analysis of these costs appears in a 2009 paper by the U.S. Department of Justice’s (DOJ) Economic Analysis Group, authored by Gerald R. Bodisch.[15] The Bodisch paper established several conservative benchmarks for understanding the costs of the traditional dealer-based distribution system.

First, Bodisch estimated that the total cost of automobile distribution in the United States averaged as much as 30% of the final vehicle price. These costs split roughly evenly between manufacturer-related expenses—such as advertising, rebates, and subsidized financing—and dealer-related expenses, including inventory financing, insurance, advertising, and sales commissions.

Second, the paper documented the scale of capital tied up in the industry’s “make-to-stock” distribution model. At the end of 2008, the nation’s roughly 20,700 franchised dealerships collectively held about $100 billion in new-vehicle inventory.[16] Bodisch estimated that carrying this inventory imposed annual costs of roughly $890 million.

Third—and most relevant for present purposes—the paper relied on a 2000 Goldman Sachs analysis of the potential savings from shifting to a direct-sales, build-to-order distribution model. Using an average vehicle price of $26,000, Goldman Sachs estimated total savings of $2,225 per vehicle, or roughly 8.6% of manufacturer’s suggested retail price (MSRP).[17] These savings fell across five categories, illustrated in Table I.

TABLE 1: Estimated Per-Vehicle Savings from Shift to Direct-Sales Model

SOURCE: Goldman Sachs (2000)

TABLE II: Range of Savings Across Multiple Price Points[18]

Goldman Sachs also identified roughly $1,000 in additional potential savings per vehicle from improvements in product development, manufacturing flexibility, and procurement. The DOJ paper attributed most of the benefits, however, to efficiencies created by shorter order-to-delivery cycles.[19]

To illustrate that these projections were not purely theoretical, the paper pointed to the Chevrolet Celta program in Brazil. There, General Motors implemented a build-to-order, internet-direct model that produced consumer prices roughly 6% lower, while selling more than 700,000 vehicles.[20]

B. Updating the DOJ Baseline

The Goldman Sachs estimates relied on data from 2000 and were reported by the DOJ in 2009. Many of the original inputs—vehicle prices, floorplan interest rates, facility costs, and inventory carrying charges—are now outdated. Importantly, most changes since 2000 have made the physical dealer infrastructure more expensive to maintain, not less.

This issue brief therefore applies the same component framework to current market conditions. The resulting figures should be understood as conservative. A fresh bottom-up reconstruction using current dealer financial data would likely produce equal or larger estimates for most components. Appendix A provides the detailed methodology, underlying data sources, and academic cross-references.

Several key inputs have shifted since the original analysis. First, the average transaction price (ATP) of a new vehicle has roughly doubled, surpassing $50,000 for the first time in September 2025.[21] Second, the interest-rate environment has tightened considerably. Dealer floorplan financing—typically structured as variable-rate facilities priced at the Secured Overnight Financing Rate (SOFR) plus 200–400 basis points—now carries effective rates of roughly 6% to 9% for well-qualified dealers, substantially higher than early-2000s levels.[22]

Other structural conditions have changed far less. Days’ supply for new vehicles still falls in the 60-90 day range identified by Bodisch, suggesting that the core inefficiencies of the make-to-stock distribution model persist despite two decades of digital transformation.[23] Dealership operating costs also remain substantial. According to the National Automobile Dealers Association (NADA), the average franchised dealership generated $73.3 million in total sales in 2024, while earning net margins of roughly 1% to 2%.[24] Those margins imply non-vehicle-acquisition operating expenses—including payroll, facilities, advertising, insurance, and floorplan interest—of about $500,000 or more per month.

Applying the Goldman Sachs component ratios to a $50,000 ATP produces the updated estimates illustrated in Table III.

TABLE III: Updated Savings Estimates for Shift to Direct-Sales Model

The updated analysis suggests that direct distribution could generate savings of roughly $3,934 to $4,992 per vehicle—up to a 9.87% efficiency gain when applied to the modern ATP.[25]

These estimates remain conservative. Elevated interest rates, with floorplan borrowing costs around 6% or higher, increase inventory carrying costs more than simple ratio scaling would suggest. Bottom-up reconstructions of individual components confirm that the ratio-based estimate likely understates the true savings. Appendix A provides the full component-level analysis, methodology, data sources, and academic cross-references.

C. Investment and Innovation Costs of Dealer Mandates

Dealer-franchise laws impose more than static distribution costs. They also create dynamic costs for manufacturing investment and innovation. Lower effective consumer prices—made possible by direct distribution—expand the addressable market for new entrants. As prices fall, more consumers can afford to purchase vehicles, increasing unit demand and improving capacity utilization at domestic production facilities. Higher volumes, in turn, make it easier for firms to justify large-scale manufacturing investments in the United States. While any estimate of incremental production is necessarily counterfactual, the underlying economic logic is straightforward: lower prices expand demand.

The innovation channel is equally important. Modern vehicles are increasingly defined by software. Over-the-air (OTA) updates, subscription-based features, and integrated digital-service ecosystems now shape how manufacturers design and deliver vehicles. These business models do not fit neatly within the mid-20th-century dealership framework—if they fit at all—which assumed that a “sale” occurred as a discrete transaction at a physical location, followed by periodic warranty or maintenance visits to a local dealer.

The tension arises because software-defined vehicles depend on an ongoing digital relationship between the manufacturer and the vehicle owner. Over-the-air (OTA) software updates allow manufacturers to remotely update vehicle systems, add features, or correct defects without requiring a visit to a dealership. These updates are delivered directly to vehicles through wireless networks, much like updates to smartphones or other connected devices.

By eliminating the need for in-person service visits, OTA updates reduce recall costs and enable continuous improvement of vehicle functionality. At the same time, they weaken the traditional dealership role as the primary intermediary for maintenance and customer contact. Dealerships derive substantial revenue from service visits and warranty work, so this shift creates a structural conflict between the software-centric design of modern vehicles and a distribution system built around periodic dealership interaction.

Manufacturers operating under a direct-sales model can instead iterate continuously on the customer experience. They can integrate vehicle purchase, software management, service scheduling, and charging infrastructure into a unified digital platform. Dealer-franchise laws often block this approach for new entrants. In many states, those statutes effectively require manufacturers to route vehicle sales through a franchised dealer network—even when no such network exists.

IV. The Dormant Commerce Clause and Dealer-Franchise Laws

The Dormant Commerce Clause is the constitutional principle that limits states’ ability to interfere with interstate commerce. Although the U.S. Constitution grants Congress the power to regulate commerce among the states, the Supreme Court has long interpreted that grant to carry a negative implication: states may not enact laws that discriminate against interstate commerce or impose burdens that outweigh their local benefits. In practice, the doctrine prevents individual states from erecting protectionist barriers or regulating commerce in ways that disrupt the national market.

State dealer-franchise laws implicate this doctrine because they dictate the business structure an out-of-state manufacturer must adopt to access a state’s consumers. These statutes require manufacturers to distribute vehicles through franchised dealers, rather than sell directly to consumers. In effect, they condition market access on adopting a state-mandated distribution model.

Direct-sales bans are vulnerable to challenge on two grounds. First, they discriminate in practical effect against out-of-state manufacturers that prefer distribution models that do not rely on in-state franchised dealers. Although these statutes appear facially neutral, they systematically advantage the incumbent dealer network—a politically powerful in-state interest group—while disadvantaging out-of-state manufacturers that seek to compete through alternative channels.

Second, even if a court determines that a direct-sales ban is not discriminatory, the statute must still survive the balancing test the Supreme Court articulated in Pike v. Bruce Church, Inc. Under that test, a state law is invalid if “the burden imposed on such commerce is clearly excessive in relation to the putative local benefits.”[26] On the burden side, the analysis is straightforward. Dealer-franchise laws impose billions of dollars in aggregate costs on consumers, fragment the national market into 50 regulatory regimes, and block the entry of competing business models. On the benefit side, the asserted local interests—consumer protection, local employment, and service infrastructure—are either pretextual or achievable through less restrictive means.[27]

Arguments that direct-sales bans protect public health or consumer welfare have repeatedly been rejected by legal scholars, economists, and staff of the Federal Trade Commission (FTC).[28] The central problem is pretext. Dealer-franchise laws were enacted to protect dealers from manufacturer coercion, not to protect consumers from manufacturers. Recasting those statutes as consumer-protection measures after the fact does not change their legislative purpose or practical effect.

Even accepting the asserted benefits at face value, categorical bans on direct sales are not narrowly tailored to achieve them. States possess a range of less restrictive alternatives. They can license direct-sales facilities, impose disclosure requirements, enforce lemon laws against all sellers regardless of distribution model, and require service-availability guarantees as a condition of market access. None of these tools requires banning an entire distribution channel.

An additional—and often overlooked—dimension of the Dormant Commerce Clause challenge involves extraterritoriality. When individual states dictate the distribution structure manufacturers must adopt, they effectively force firms to organize their national operations around the most restrictive state’s rules. Manufacturers cannot realistically maintain one distribution system for permissive states and another for restrictive states without incurring large duplicative costs. The result is that a handful of restrictive states determine the distribution architecture for the entire national market—precisely the kind of extraterritorial regulation the Commerce Clause was designed to prevent.[29]

Courts and policymakers have begun to grapple with these tensions. In Massachusetts, the Supreme Judicial Court held that the state’s dealer association lacked standing to challenge Tesla’s direct-sales operations because Tesla had no affiliated dealer network. The court recognized that Tesla’s business model presented no franchise relationship to regulate.[30] In Illinois, an appellate court ruled in 2024 that existing law does not bar manufacturers from obtaining dealer licenses to conduct direct sales.[31]

Manufacturers continue to test these legal boundaries. Lucid has filed challenges in multiple states.[32] Rivian has backed reform efforts in Washington state, including a now-abandoned 2026 ballot initiative to authorize direct-to-consumer EV sales.[33] Rivian has also filed suit in Ohio challenging a state direct-sales ban that permits Tesla to sell vehicles directly while excluding newer entrants.[34]

Taken together, these developments suggest that courts and legislatures[35] are beginning to recognize a key distinction: regulating existing franchise relationships differs fundamentally from prohibiting alternative business models altogether.

V. Statutory Misfit in the Modern Vehicle Market

Even before reaching constitutional questions, many state franchise statutes fit poorly with modern vehicle business models. Courts and regulators often apply these laws to direct-to-consumer manufacturers through expansive interpretations of provisions designed for a fundamentally different commercial relationship.

Most franchise statutes revolve around the manufacturer-dealer relationship. Their operative provisions govern franchise termination, territorial allocations, facility requirements, and other features of an ongoing vertical business arrangement. Direct-to-consumer entrants have none of these elements. They operate without franchisees, franchise agreements, or territorial allocations. Applying franchise statutes to such firms does not advance the statutes’ original purpose. Instead, it stretches the laws beyond their text and structure to reach conduct the legislature never contemplated.[36]

The modern automotive market now includes business models that diverge sharply from the franchised-dealer archetype. Manufacturers increasingly rely on online ordering with centralized, non-negotiable pricing; agency sales models in which the “dealer” functions primarily as a delivery agent, rather than a reseller; remote delivery and mobile-service routing; software-defined vehicle features delivered through over-the-air updates; vertically integrated charging and service networks; and hybrid retail models in which showrooms exist but do not operate as statutory “dealers.” These approaches challenge the core assumptions of franchise law, which presupposed a manufacturer producing a finished vehicle, shipping it to a dealer’s lot, and allowing the dealer to negotiate the retail price with the consumer.

State regulatory approaches to these models vary widely. In practice, state regimes generally fall into four categories: outright bans on direct manufacturer sales; bans with limited electric-vehicle carve-outs; capped regimes that permit direct sales subject to limits on the number of stores or vehicles sold; and permissive regimes that allow direct sales with few restrictions. This patchwork fragments the national market and complicates manufacturers’ efforts to operate under a consistent distribution model.

In many restrictive states, regulators stretch the statutory definitions of “dealer” and “franchise” beyond recognition to capture manufacturers that have never operated through a franchise model. This interpretive expansion is not merely technical. It determines whether new competitors can reach millions of consumers.

It also bears noting that federal law already addresses many of the concerns that state franchise statutes claim to address. Federal dealer-protection statutes govern manufacturer-dealer relationships nationwide.[37] The FTC has also attempted to regulate dealer practices through rulemaking. Although a federal court vacated the agency’s proposed CARS Rule on procedural grounds,[38] the rule illustrates that federal regulators can pursue consumer-protection objectives through less restrictive tools than blanket bans on direct distribution. The presence of these federal mechanisms weakens the argument that state-level distribution bans are necessary to protect consumers.

VI. Conclusion

State automobile-dealer franchise laws impose what this issue brief calls a “middleman tax” on American car buyers. Applying the DOJ’s methodology with current market data, the analysis here estimates that mandatory dealer intermediation adds roughly $4,687 to the cost of a new vehicle. That figure rises further when dealer add-ons, bargaining rents, and administrative fees are included. In a market where the average transaction price exceeds $50,000 and affordability sits near historic lows, the issue is not abstract. It directly affects millions of consumers.

These estimates are conservative. Even using the lowest plausible inputs for each component, the floor estimate remains about $3,934 per vehicle, or 7.87% of the average transaction price. The increase in the average transaction price—from $26,000 in 2000 to about $50,000 today—largely tracks cumulative inflation. That means the structural cost of mandatory intermediation remains at least as large in real terms today as when the DOJ first documented the problem more than 15 years ago. Current market conditions likely increase the burden further. Higher interest rates, rising commercial real-estate costs, and greater vehicle complexity all raise the cost of maintaining a large brick-and-mortar dealer network. Bottom-up reconstructions using current dealer financial data consistently match or exceed the ratio-scaled estimates. A realistic estimate places the cost closer to $4,992 per vehicle, even before accounting for bargaining frictions, dealer add-ons, and consumer time costs documented in the academic literature.

Dealer franchise laws also impose broader economic costs. They fragment the national market by forcing manufacturers to organize distribution around the most restrictive state rules. They burden interstate commerce by conditioning market access on a state-mandated business structure. And they fit poorly with modern vehicle business models, extending franchise-protection statutes designed for mid-20th-century manufacturer-dealer relationships to firms that have never operated through a franchise system.

The principle guiding reform should be simple: allow channel competition. States should not mandate a single distribution architecture when multiple models can compete to serve consumers.

Several reform paths are available. The most comprehensive option is repeal of direct-sales bans, allowing manufacturers to choose any lawful distribution model. Where repeal proves politically infeasible, states can replace categorical bans with licensing and disclosure requirements that ensure consumer protection without prohibiting direct sales. States can also require service availability, bonding, or escrow provisions as conditions of market access. Any reform should include sunset clauses and periodic review requirements to reduce the risk of regulatory recapture by incumbent interests.[39]

The federal government also has a role to play. The DOJ’s Anticompetitive Regulations Task Force could issue a formal report updating the 2009 Bodisch analysis with current market data and documenting the competitive harms of state direct-sales bans. The DOJ could also file statements of interest or amicus curiae briefs in ongoing litigation challenging these laws. In cases where state statutes conflict with the Dormant Commerce Clause or federal policy, the federal government may also consider direct legal challenges.

Congress also has a role to play. For example, it could structure relevant federal incentive programs to encourage states to adopt more forward-thinking direct-sales policies.

The case for reform is strong. Protecting an incumbent distribution channel is not the same as protecting consumers. Allowing manufacturers to compete through different distribution models would lower costs, expand consumer choice, and better align the law with the realities of the modern automobile market.

APPENDIX A: Methodology and Data Sources for the Reconstructed Goldman Sachs Distribution Savings Decomposition

I. Methodology and Baseline Assumptions

This appendix documents the methodology, data inputs, and academic cross-references underlying the updated distribution-cost estimates presented in Section III.B of the issue brief. The analytical framework originates in a 2000 Goldman Sachs equity-research report. That report estimated that shifting from the franchised-dealer distribution model to a direct-sales, build-to-order system could reduce distribution costs by roughly $2,225 per vehicle, or 8.56% of the then-average transaction price of $26,000.[40] The DOJ’s Economic Analysis Group later adopted this framework in a 2009 discussion paper to demonstrate that state dealer-franchise laws impose measurable efficiency costs on automobile distribution.[41]

The Goldman Sachs analysis relied on economic conditions that differ substantially from today’s market. Many of the original inputs—including floorplan financing rates, facility costs, and inventory-carrying charges—reflect early-2000s conditions. Over the past quarter-century, rising vehicle prices, fluctuating interest rates,[42] and higher commercial real-estate costs[43] have altered the cost structure of dealer-based distribution. These changes generally increase, rather than reduce, the cost of maintaining a large physical dealership network.

The reconstruction presented here applies the original percentage decomposition to updated market data from 2025–2026. This approach provides a conservative estimate. A fully bottom-up reconstruction using current dealer financial data would likely generate equal or larger estimates for most cost components.

The analysis uses an average transaction price (ATP) of $50,000 as the primary baseline. This figure reflects the September 2025 record of $50,080 reported by Cox Automotive’s Kelley Blue Book.[44] Each component estimate presented below draws on current industry data and, where available, cross-references the academic literature summarized in the accompanying literature review.

II. Original Component Decomposition

The 2000 Goldman Sachs report decomposed potential distribution-cost savings into five categories, each expressed as a percentage of the average vehicle price.[45] The DOJ’s Economic Analysis Group adopted these estimates in its 2009 analysis, noting that total distribution costs could reach as much as 30% of the final vehicle price, split roughly evenly between manufacturer-side and dealer-side expenses.[46]

APPENDIX TABLE I: Goldman Sachs 2000 Baseline

SOURCE: Goldman Sachs (2000)

The Goldman Sachs report also identified roughly $1,000 per vehicle in additional savings from secondary sources, including modular vehicle architecture and tighter supplier integration. Those effects are not included in the five-component decomposition reported above.[47]

III. Updated Macroeconomic and Industry Inputs

The reconstruction requires updating the key macroeconomic and industry variables that drive each savings component. The table below summarizes the baseline and updated values:

Updating the original framework requires revising the macroeconomic and industry variables that drive each savings component. Table II below summarizes the baseline values used in the 2000/2009 analysis and the updated inputs used in the reconstruction.

APPENDIX TABLE II: Updated Economic Inputs (2025–2026)

[48] [49] [50] [51] [52] [53] [54] [55] [56]

A. Note on ATP Selection

This analysis uses $50,000 as a round-number proxy for the modern average transaction price (ATP). Cox Automotive reported an ATP of $50,080 in September 2025,[57] while the December 2025 figure was approximately $47,104 for all light vehicles.[58] MoneyGeek projects an average ATP of $48,841 for 2026.[59] The $50,000 baseline therefore sits near the upper end of recent estimates. Sensitivity to ATP assumptions is addressed in Section VI below.

IV. Component-by-Component Reconstruction

This section reconstructs each component of the Goldman Sachs distribution-savings framework using updated market data. Subsection A examines savings from improved matching of production to consumer demand under a build-to-order or direct-sales model. Subsection B estimates reductions in inventory-carrying costs resulting from shorter supply chains and lower days’ supply. Subsection C analyzes potential savings from reducing dealership density and the fixed facility overhead associated with large retail networks. Subsection D evaluates reductions in sales commissions and personnel costs that arise when negotiated dealership sales processes are replaced with centralized, non-negotiable pricing. Subsection E estimates potential logistics savings from streamlined hub-to-consumer delivery and reduced dealer-network routing, and concludes with an empirical cross-check using Tesla’s operating experience as a direct-sales manufacturer. Together, these components provide a conservative reconstruction of the distribution-cost savings discussed in Section III.B of the issue brief.

A. Component 1: Improved Matching of Supply with Consumer Demand

Original Ratio: 3.20% of vehicle price.[60]

Mechanism: Under the make-to-stock model, dealers must clear unpopular configurations through incentives and discounts. A build-to-order (BTO) or direct-sales model reduces the need for such price-clearing mechanisms by aligning production with confirmed consumer preferences. The Goldman Sachs framework attributed the largest share of projected savings to this channel.[61]

Calculation:

S? = ATP × 0.032

S? = $50,000 × 0.032 = $1,600

1. Supporting data

Manufacturer incentive spending reached roughly 7.5% of ATP (about $3,750 per vehicle) by late 2025,[62] indicating that the costs of mismatched inventory remain substantial. Academic research corroborates that search and matching frictions generate meaningful price distortions. Charles Murry and Yiyi Zhou estimate that search frictions create markups of roughly $333 per vehicle in markets where dealers collocate.[63] That figure captures only consumer-side search costs, not the manufacturer-side incentives required to move mismatched inventory. Florian Zettelmeyer, Fiona Scott Morton, and Jorge Silva-Risso find that internet-based information reduces transaction prices by roughly 1.5%–2%, largely through improved consumer matching and bargaining leverage.[64] These results support the core mechanism in the Goldman Sachs framework.

2. Assessment

The 3.2% ratio remains plausible and may even be conservative under current market conditions. In 2000, incentive spending represented a smaller share of vehicle prices. By 2025, manufacturer incentives reached about 7.5% of ATP (roughly $3,750 per vehicle),[65] far exceeding the $832 the original model attributed to supply-demand mismatching at a $26,000 ATP. This pattern suggests that inefficiencies associated with make-to-stock distribution have grown relative to the original estimate. The $1,600 figure should therefore be understood as the maximum savings achievable under full BTO adoption. Even a partial shift toward build-to-order distribution could capture a meaningful share of these gains.

B. Component 2: Lower Inventory Carrying Costs

Original Ratio: 2.21% of vehicle price.[66]

Mechanism: The franchised-dealer model requires maintaining roughly 60–90 days’ supply of finished vehicles on dealer lots, generating substantial floorplan interest and operational holding costs. Direct-sales models can reduce this inventory buffer by aligning production more closely with confirmed consumer demand.[67]

Ratio-Based Calculation:

S? (ratio) = ATP × 0.0221

S? = $50,000 × 0.0221 = $1,105

Bottom-Up Reconstruction: Because the full Goldman Sachs report is not publicly available, this appendix also reconstructs inventory savings using current data inputs. Inventory-carrying costs include both an interest component and a noninterest operational component:

S? (reconstructed) = (ATP × r × Δt / 365) + (Δt × C??)

Where:

• r ≈ 6% (approximate floorplan rate, conservative relative to current prime-plus dealer financing);[68]
• Δt = 45 days (estimated reduction from a 73-day legacy supply to roughly a 28-day direct-to-consumer target);[69] and
• C?? = $15 per day (noninterest holding costs, including insurance, storage, and lot maintenance).[70]

S? (reconstructed) = ($50,000 × 0.06 × 45 / 365) + (45 × $15)

= $370 + $675

= $1,045

The bottom-up reconstruction yields $1,045 (2.09% of ATP), slightly below the $1,105 (2.21%) produced by ratio scaling. The true value likely falls within a range of roughly $1,045–$1,105 per vehicle, or about 2.09%–2.21% of the vehicle price. Recent industry data show net floorplan expense per unit rising 35% to $487 in the first quarter of 2025, reflecting the upward pressure of higher interest rates. The 6% floorplan rate assumed here is conservative. With prime rates between roughly 6.5% and 7.5% over the past year,[71] actual dealer floorplan rates often approach or exceed 7%, which would increase the reconstructed estimate.

1. Academic cross-check

Empirical research supports the importance of the inventory channel. Gérard P. Cachon and Marcelo Olivares show that distribution-network design strongly influences finished-goods inventory levels and that inter-dealer competition encourages excess stocking.[72] Their findings support the assumption that a direct-sales model with fewer distribution points would reduce days’ supply. Adam Copeland, Wendy E. Dunn, and George Hall document systematic within-model-year price declines under the build-to-stock model, as dealers discount aging inventory to clear lots.[73]

2. Assessment

The bottom-up reconstruction yields $1,045 (2.09% of ATP), modestly below the ratio-scaled estimate of $1,105 (2.21%). Both reconstruction inputs are deliberately conservative. The assumed 6% floorplan rate sits below current prime-based dealer financing rates, which often fall in the 7%–10% range. The $15-per-day noninterest holding cost also represents a lower-bound estimate derived from aggregate dealership operating data. Adjusting either input to reflect current market conditions would increase the reconstructed value. For consistency with the original Goldman Sachs framework, this brief uses the ratio-based figure ($1,105) in the summary table. That figure falls comfortably within the range supported by current data and likely understates inventory-carrying costs in the present interest-rate environment.

C. Component 3: Reduction in Dealership Density and Facility Overhead

Original Ratio: 1.49% of vehicle price.[74]

Mechanism: Physical dealerships impose large fixed costs, including facility leases or mortgages, insurance, utilities, maintenance, and administrative personnel. A direct-sales model replaces dispersed suburban dealership lots with centralized delivery hubs, service centers, and smaller retail galleries, reducing the need for a dense network of full-service retail locations.

Ratio-Based Calculation:

S? (ratio) = ATP × 0.0149

S? = $50,000 × 0.0149 = $745

Bottom-Up Reconstruction: This component can also be reconstructed using current dealership-level data. As of 2024, the United States had approximately 16,957 franchised light-vehicle dealerships.[75] NADA reports that the average franchised dealership generated $73.3 million in total sales in 2024 on net margins of approximately 1–2 percent, with cost of goods sold (vehicle acquisition) representing roughly 90 percent of revenue. This implies non-vehicle-acquisition operating expenses on the order of $500,000 or more per month per dealership, encompassing payroll, facility costs, advertising, insurance, and floorplan interest.

Note on Cost Allocation: These operating expenses include both facility-related overhead (leases or mortgages, utilities, property insurance, taxes, building maintenance, and management) and personnel-related costs (sales commissions, salaries, and benefits). Because the Goldman Sachs framework treats commissions separately, as in Component 4, this analysis allocates only a portion of operating expenses to facility overhead to avoid double counting.

No single authoritative source reports the facility-related share of dealership operating costs. This appendix estimates facility-related overhead at roughly $100,000–$125,000 per month per dealership, covering lease or mortgage payments, utilities, insurance, property taxes, and facility maintenance. This assumption is conservative. A typical multiacre suburban dealership with showroom space, service bays, and large vehicle lots likely incurs facility costs at or above this range, particularly given commercial real-estate appreciation since 2000. Even the lower bound of $100,000 represents only about 20% of total estimated monthly operating expenses.

Calculation (Facility-Related Overhead, Assuming a 75% Network Reduction):

Total annual facility overhead:

16,957 dealerships × $112,500 per month (midpoint) × 12 months

= $22.9 billion per year

Per vehicle sold (14.9 million seasonally adjusted annual rate, or SAAR)[76]:

$22.9 billion ÷ 14.9 million = $1,536 per vehicle

Assuming a 75% reduction in the physical retail network under a direct-sales model:

S? (reconstructed) = $1,536 × 0.75 = $1,152 (2.3% of ATP)

For comparison, if facility-related overhead were as low as $50,000 per month—capturing only lease or mortgage costs and excluding other facility expenses—the floor estimate would be:

16,957 × $50,000 × 12 ÷ 14.9 million × 0.75

= $512 per vehicle (1.02% of ATP)

The bottom-up reconstruction therefore produces a range of $512–$1,152 per vehicle. The broader facility-overhead estimate ($1,152, or 2.3%) exceeds the ratio-scaled Goldman Sachs estimate of $745 (1.49%). The Goldman Sachs figure falls near the midpoint of the reconstructed range, suggesting that it captured more than bare lease costs but less than the full allocation of facility overhead.

A 75% reduction in the retail network is itself conservative relative to observed direct-sales models. Tesla serves U.S. sales volumes comparable to those of a mid-size franchised manufacturer through roughly 276 retail locations, whereas a similarly sized franchised brand typically operates between about 1,200 and 4,000 dealerships. That comparison implies network reductions of roughly 77%–93% on a per-brand basis. This appendix uses the lower figure to reflect that a mature direct-sales system would still require delivery hubs, service centers, and test-drive locations.

1. Academic cross-check

Empirical research on selling, general, and administrative (SG&A) costs shows that these expenses often exhibit “stickiness,” adjusting slowly when demand declines.[77] This literature suggests that overhead embedded in a mandated intermediary structure can persist even when more efficient configurations become available.[78] Scholarship on franchise law similarly characterizes dealership infrastructure as an institutional artifact: when intermediation is mandatory and entry and exit are constrained, the physical retail network persists even if market conditions would otherwise favor consolidation or alternative distribution models.

2. Assessment

The bottom-up analysis brackets the Goldman Sachs estimate. A conservative lease-only assumption supports savings of about $512 per vehicle (1.0%), while a broader facility-overhead allocation supports roughly $1,152 per vehicle (2.3%). The original Goldman Sachs ratio of 1.49% ($745) falls comfortably within this range and serves as a reasonable midpoint estimate. This appendix therefore adopts the Goldman Sachs ratio as the lower-bound estimate. Changes in underlying cost drivers since 2000 reinforce that conclusion. Commercial real-estate prices, property insurance, and utility costs have all risen substantially, increasing the per-unit cost of maintaining large suburban dealership facilities relative to the environment in which the original estimate was developed.

D. Component 4: Lower Sales Commissions and Personnel Expenses

Original Ratio: 1.47% of vehicle price.[79]

Mechanism: The franchised-dealer sales process relies on multiple layers of commissioned and salaried personnel: front-line salespeople, sales managers who approve and negotiate deals, finance-and-insurance (F&I) managers who structure financing and sell add-on products, and administrative staff who support the transaction. A direct-sales model replaces this structure with salaried product specialists operating under centralized, non-negotiable pricing. That shift reduces the multi-layered personnel overhead associated with commission-based bargaining.

Ratio-Based Calculation:

S? (ratio) = ATP × 0.0147

S? = $50,000 × 0.0147 = $735

Partial Bottom-Up Reconstruction: Average gross profit per new vehicle retailed was $2,247 in 2024. Industry-standard commission rates range from about 20%–25% of front-end gross profit.[80] Applying that range to the reported gross profit yields a direct per-vehicle commission of roughly $449–$562,[81] establishing a verifiable floor for the commission component (about 0.90%–1.12% of ATP).

Front-line commissions represent only part of the sales-department personnel cost that a direct-sales model would restructure. The remainder of the Goldman Sachs estimate—about $286 per vehicle (the difference between $735 and the lower-bound commission of $449)—likely reflects additional personnel layers involved in the traditional dealership sales process, including:

• Sales-manager compensation (deal negotiation, desk approval, and pricing authority)
• F&I-manager compensation (financing structuring, product sales, and compliance)
• Administrative and desk staff supporting the multi-step transaction process
• Training and onboarding costs for a high-turnover commissioned sales force

No single source reports per-vehicle figures for each of these components. A conservative allocation nonetheless aligns with the Goldman Sachs total: $449 in direct commissions plus roughly $286 in supporting sales-department personnel costs equals $735 (1.47% of ATP). This estimate implies that for every dollar paid in front-line commission, about $0.64 supports the surrounding sales infrastructure—a plausible ratio given the management-intensive and negotiation-driven nature of the franchised-dealer sales process.

1. Academic cross-check

Academic research documents the welfare costs associated with commission-driven bargaining models. Ambarish Chandra, Sumeet Gulati, and James Sallee find that negotiated pricing disproportionately disadvantages consumers with higher bargaining disutility, including many older buyers and women.[82] In effect, bargaining can function as a regressive transaction tax paid through time, stress, and higher expected prices. Meghan R. Busse, Jorge Silva-Risso, and Florian Zettelmeyer show that consumer-facing rebates pass through more directly to buyers than dealer-targeted incentives, suggesting that intermediaries capture surplus through opaque pricing channels.[83] Andreas Grunewald et al. estimate that eliminating dealer discretion in loan pricing alone would generate significant consumer-surplus gains—an additional cost channel beyond the commission savings estimated here.[84]

2. Assessment

The direct commission estimate of $449–$562 per vehicle provides a well-supported lower bound. The Goldman Sachs ratio of 1.47% ($735) plausibly captures the full sales-department personnel structure, although the portion above the direct commission (about $286) reflects supporting infrastructure costs that cannot be independently verified with the same precision. Some portion of this estimate may overlap with personnel costs embedded in dealership overhead (Component 3). That overlap is likely minimal because the bottom-up reconstruction in Component 3 relies primarily on facility and occupancy costs, rather than staffing figures.

The $735 estimate therefore remains a reasonable midpoint for the personnel-cost component. Importantly, it captures only the direct payroll effects of restructuring the sales process. It does not include broader welfare gains from eliminating bargaining frictions—such as reduced consumer time costs, lower price dispersion, and improved transparency—which the academic literature suggests could be substantially larger than the commission savings alone.

E. Component 5: Lower Shipping and Logistics Costs

Original Ratio: 0.19% of vehicle price.[85]

Mechanism: Direct-sales models enable optimized hub-to-consumer delivery routes, reducing reliance on the multi-step manufacturer-to-dealer-to-consumer logistics chain that characterizes the franchised-dealer system. Modern destination fees have also increased substantially, with some domestic trucks carrying destination charges as high as $2,295.[86]

Calculation:

S? = ATP × 0.0019

S? = $50,000 × 0.0019 = $95

1. Assessment

This component is the smallest in the Goldman Sachs framework and likely the most conservative. Current destination fees—generally ranging from $995 to $2,295—substantially exceed the $95 savings estimated here. That gap suggests the original Goldman Sachs ratio may understate potential logistics savings. Destination fees, however, include transportation costs from the assembly plant to the retail network that would exist under any distribution model. The marginal savings from logistics optimization therefore remain modest relative to the other components.

At the same time, several modern logistics factors suggest that the $95 estimate likely understates the full inefficiency associated with dealer-based routing.

First, destination fees have risen sharply—from an average of about $839 in 2011 to a current range of roughly $995–$2,295, with a midpoint near $1,200. These fees are typically equalized across buyers of a given model regardless of distance from the assembly plant. That structure effectively cross-subsidizes transportation costs and reflects the logistical constraints of routing vehicles through a dispersed network of roughly 17,000 dealer lots, rather than through a smaller number of regional delivery hubs. A hub-based distribution system could price logistics more directly based on distance. If even 15%–25% of the average destination fee reflects this structural inefficiency, the avoidable cost would fall in the range of roughly $180–$300 per vehicle.^[87]

Second, dealer-to-dealer trades—in which a vehicle is transported from one dealership to another to match a buyer’s desired configuration—represent pure logistics waste that would not exist under a build-to-order or centralized-inventory model. Transfer fees for these trades range from $0 for short intra-market swaps to several hundred dollars for cross-state transports, with documented examples around $195 per transaction. Although no authoritative source quantifies the share of transactions involving dealer trades, the practice is common enough to support a dedicated logistics infrastructure.^[88]

Third, direct-to-consumer operators demonstrate that alternative logistics systems can operate at scale. Carvana, for example, runs a hub-and-spoke distribution network with vertically integrated transportation, reconditioning, and customer delivery. The company reports per-unit cost advantages of roughly $1,200–$2,600 relative to traditional dealer-intermediated distribution.^[89]

Taken together, these factors suggest that a reasonable but nonconservative estimate of logistics savings would fall in the range of $200–$400 per vehicle (roughly 0.40%–0.80% of ATP), compared with the $95 estimate produced by ratio scaling. This appendix nonetheless retains the $95 figure as the baseline, because the original Goldman Sachs derivation is not publicly available in sufficient detail to determine which logistics costs it already incorporated. The $200–$400 range should therefore be understood as a plausible upper bound reflecting destination-fee inefficiencies, dealer-trade logistics waste, and the demonstrated efficiency of hub-based direct-delivery models.

2. Cross-check

An additional cross-check for the savings estimate comes from Tesla’s operating experience as a direct-sales manufacturer. Bloomberg Intelligence reports that Tesla’s SG&A expense per delivery was $2,651 per vehicle in 2023, about 72% higher than Stellantis.[90] This figure reflects the cost of maintaining company-owned showrooms, mobile-service crews, and vertically integrated delivery logistics.

These figures provide a practical benchmark against the theoretical savings estimate, demonstrated in Table III. The resulting net efficiency gain of roughly $2,036 per vehicle indicates that even after accounting for the substantial costs of operating a vertically integrated retail and service network, direct sales can remain more cost-efficient than the franchised-dealer model.

APPENDIX TABLE III: Tesla Benchmark Against Theoretical Savings

[91]

The $2,651 figure also reflects Tesla’s continued buildout of its retail, service, and delivery infrastructure. As the network matures, economies of scale could reduce SG&A costs on a per-vehicle basis.

For policy purposes, the key question is not whether direct sales always produce lower costs in every circumstance. The relevant question is whether state law should categorically prohibit manufacturers from attempting a distribution model that economic evidence suggests can produce measurable efficiencies. Even the $2,036 net estimate represents a meaningful cost reduction for consumers.

V. Methodological Limitations and Caveats

The reconstruction presented in this appendix provides a conservative estimate of the potential distribution-cost savings from direct-sales models, but several methodological limitations should guide how the results are interpreted. The analysis relies on ratio scaling from the original Goldman Sachs framework, rather than a fully bottom-up re-derivation using current dealer cost data. It also evaluates the distribution channel in partial equilibrium, holding broader market conditions constant. In addition, structural changes in the automotive industry over the past quarter-century—including dealer consolidation, digital retail tools, and changes in product mix—may affect how the original ratios translate to current market conditions. The subsections below discuss these limitations and explain why, despite them, the estimates presented here likely understate rather than overstate the cost of mandatory dealer intermediation.

A. Linear Scaling of Cost Ratios

The methodology applies the original percentage ratios from the 2000 Goldman Sachs analysis to modern vehicle prices, implicitly assuming that the cost components scale proportionally with the average transaction price. In practice, some components may scale nonlinearly. Logistics costs, for example, may include larger fixed components, while savings from improved supply-demand matching could increase as vehicles become more complex. The directional bias of this assumption likely produces understatement. Over the past 25 years, interest rates, commercial real-estate costs, and vehicle complexity have all increased, factors that would tend to raise the cost of maintaining a large dealership network. A fully bottom-up reconstruction using current cost structures would therefore likely produce estimates at least as large as the ratio-scaled figures presented here.

B. Industry Changes Since 2000

The automotive industry has changed significantly since the Goldman Sachs report. Dealer consolidation may have produced some efficiency gains within the franchise system. Digital shopping tools have reduced some consumer information frictions. At the same time, the shift in product mix toward larger vehicles, particularly SUVs and trucks, has altered cost structures across manufacturing and distribution. These structural changes could influence the magnitude of distribution-cost savings in either direction.

C. Original Ratios Not Re-Derived

The 2026 estimates apply the original Goldman Sachs percentage decomposition to updated price data, rather than independently re-estimating each component using current cost structures. Future research should attempt a full bottom-up reconstruction using contemporary dealer financial data, such as the National Automobile Dealers Association’s “Annual Financial Profile of America’s Franchised New-Car Dealerships,” rather than relying solely on ratio extrapolation.

D. Partial-Equilibrium Framework

The estimates describe the savings associated with transitioning a single manufacturer’s distribution model while holding other market conditions constant. A system-wide transition to direct distribution could generate general-equilibrium effects, including changes in competitive dynamics, entry and exit patterns, and manufacturer pricing strategies. These second-order effects could be substantial and are ambiguous in direction.

E. Inflation Adjustment

The reconstruction applies the original Goldman Sachs percentage ratios to the 2026 average transaction price of $50,000 rather than simply inflating the original dollar figures from 2000. This approach is appropriate because cumulative consumer price index inflation from 2000 to 2025 is roughly 92%.[92] The increase in the average transaction price from $26,000 to about $50,000 therefore largely reflects general inflation, with only modest real appreciation. Applying the original ratios to the modern transaction price produces estimates broadly equivalent to an inflation-adjusted calculation.

VI. Conclusion

The reconstruction presented in this appendix is deliberately conservative at every step. It applies simple ratio scaling from the original Goldman Sachs framework, adopts the lowest plausible inputs where ranges exist, and explicitly flags areas of uncertainty. Under these cautious assumptions, the floor estimate is $3,934 per vehicle (7.87% of ATP), derived by taking the lowest value in each component range. Because cumulative CPI inflation from 2000 to 2025 is roughly 92%, the increase in the average transaction price from $26,000 to $50,000 largely reflects general price-level changes. Even the conservative floor therefore implies that the real cost of mandatory dealer intermediation is at least as large today as when Goldman Sachs first estimated it in 2000 and the DOJ documented it in 2009.

The conservative floor, however, is unlikely to represent the most realistic estimate. The bottom-up evidence consistently points in one direction: current market conditions have made the physical dealer infrastructure more expensive to maintain in real terms, not less. Interest rates are higher than in 2000, commercial real estate costs have risen, and vehicle complexity has increased. Where components could be reconstructed using contemporary data, the bottom-up figures generally met or exceeded the ratio-scaled estimates—inventory carrying costs of $1,045–$1,105, facility overhead of $745–$1,152, commissions of $449–$735, and shipping and logistics savings of up to $400 per vehicle. Using the upper end of these component ranges yields an estimate of roughly $4,992 per vehicle (9.87% of ATP). Even that figure likely understates the full economic burden of mandatory dealer intermediation because it excludes welfare costs—such as bargaining frictions, consumer time costs, and distributional harms—that the academic literature identifies as economically significant but that fall outside the Goldman Sachs accounting framework.

[1] See Daniel A. Crane, Tesla, Dealer Franchise Laws, and the Politics of Crony Capitalism, 101 Iowa L. Rev. 573 (2016),  https://repository.law.umich.edu/articles/1721.

[2] These figures represent the per-vehicle cost based on an average purchase price of $50,000, as discussed infra.

[3] See Gerald R. Bodisch, Economic Effects of State Bans on Direct Manufacturer Sales to Car Buyers, U.S. Dep’t Just. (2009), https://www.justice.gov/atr/economic-effects-state-bans-direct-manufacturer-sales-car-buyers.

[4] See Cox Auto. Inc., Kelley Blue Book Report: New-Vehicle Average Transaction Price Hits Record High in September, Surges Past $50,000 for the First Time Ever (Oct. 13, 2025), https://www.coxautoinc.com/insights-hub/sept-2025-atp-report.

[5] See Jonathan Gregory, Wage Gains Boost New-Vehicle Affordability in November Despite Higher Rates, Cox Auto. Inc. (Dec. 15, 2025), https://www.coxautoinc.com/insights-hub/nov-2025-vai (noting that purchasing a new vehicle requires about 36.3 weeks of income—still historically high, despite recent affordability improvements following pandemic-driven price increases).

[6] See Roger D. Blair et al., Brief of Legal and Economic Scholars to the Georgia Supreme Court in Lucid v. Georgia, Int’l Ctr. L. & Econ. (July 10, 2025), https://laweconcenter.org/resources/brief-of-legal-and-economic-scholars-to-the-georgia-supreme-court-in-lucid-v-georgia.

[7] Id.

[8] James M. Rubenstein, Making and Selling Cars: Innovation and Change in the U.S. Automotive Industry 188 (Johns Hopkins Univ. Press 2008).

[9] See Patrick Manzi, 2024 Annual Financial Profile of America’s Franchised New-Car Dealerships, Nat’l Auto. Dealers Ass’n (2024), https://www.nada.org/media/4695/download.

[10] Julie Walker, How Many U.S. Dealership Owners Will We Have in 2050?, Kerrigan Advisors (Mar. 30, 2025), https://www.kerriganadvisors.com/in-the-news/how-many-u-s-dealership-owners-will-we-have-in-2050.

[11] Mass. State Auto. Dealers Ass’n, Inc. v. Tesla Motors MA, Inc., 15 N.E.3d 1152 (Mass. 2014), https://law.justia.com/cases/massachusetts/supreme-court/2014/sjc-11545.html.

[12] See George J. Stigler, The Theory of Economic Regulation, 2 Bell J. Econ. & Mgmt. Sci. 3 (1971).

[13] Bodisch, supra note 3.

[14] 2024 Car Buyer Journey Study, Cox Auto. Inc. (2025), https://www.coxautoinc.com/wp-content/uploads/2025/01/2024-Car-Buyer-Journey-Study-Research-Summary.pdf.

[15] Id.

[16] Id. at 4.

[17] Id.

[18] As discussed below, this analysis yields estimated savings ranging from a conservative baseline of 7.87% per vehicle to 9.87% per vehicle. The analysis assumes a $50,000 new-vehicle price. Table II illustrates how these estimates translate across a range of vehicle price points.

[19] Id.

[20] Id. at 5-6.

[21] Cox Auto Inc., supra note 4 (noting that recent increases in average vehicle prices largely track overall inflation; see also Appendix A, § 7.E); Nora Eckert, Automakers Have Resisted Raising Car Prices Because of Tariffs. That Might Not Last, Reuters (Sept. 18, 2025), https://www.reuters.com/business/media-telecom/automakers-have-resisted-raising-car-prices-because-tariffs-that-might-not-last-2025-09-18 (reporting that 2025 U.S. tariffs—including 25% duties on imported vehicles and auto parts—have increased costs for automakers and suppliers, potentially adding about $2,300 per vehicle); The Latest Car Tariff Information, Kelley Blue Book (Feb. 25, 2026), https://www.kbb.com/tariffs/#:~:text=New%20cars%3A%20Shoppers%20can%20expect,cars%20less%20affordable%20for%20consumers (estimating that tariffs could raise prices by as much as $6,000 on vehicles priced under $40,000 as duties phase in); Lawrence Ulrich, Trump’s Auto Tariffs Are Turning New-Car Shopping into a Wild Ride, Car & Driver (Aug. 5, 2025), https://www.caranddriver.com/news/a65592759/trump-auto-tariffs-new-car-shopping-outlook.

[22] See Thomas L. Hidder, Floor-Plan Financing for Auto Dealers: Trends, Structures & What’s Changing, Harney Partners (Nov. 24, 2025), https://harneypartners.com/floor-plan-financing-for-auto-dealers; Prime Rate Information, Bank Am. (Dec. 11, 2025), https://newsroom.bankofamerica.com/content/newsroom/home/prime-rate-information.html (reporting a prime rate of 6.75% effective December 2025; this brief uses an approximate 6% rate in the bottom-up reconstruction as a conservative assumption; see Appendix A).

[23] See Cox Auto. Inc., New-Vehicle Inventory Returns to Pre-Tariff Levels as EV Sales Accelerate and Incentives Rise (Aug. 14, 2025), https://www.coxautoinc.com/insights-hub/july-2025-new-vehicle-inventory.

[24] See Appendix A § 5.

[25] See Optimum Info, Dealer Financial Analysis Report—Q1 2025, https://optimuminfo.com/resources/blogs/dealer-financial-analysis-report-q1-2025 (last visited Feb. 27, 2025).

[26] Pike v. Bruce Church, Inc., 397 U.S. 137, 142 (1970).

[27] See Fed. Trade Comm’n, FTC Staff: Missouri and New Jersey Should Repeal Their Prohibitions on Direct-to-Consumer Auto Sales by Manufacturers (May 16, 2014), https://www.ftc.gov/news-events/news/press-releases/2014/05/ftc-staff-missouri-new-jersey-should-repeal-their-prohibitions-direct-consumer-auto-sales.

[28] Id.

[29] See Healy v. Beer Inst., 491 U.S. 324, 336 (1989) (“The Commerce Clause … precludes the application of a state statute to commerce that takes place wholly outside of the State’s borders, whether or not the commerce has effects within the State.”); Crane, supra note 1 (arguing that state dealer-franchise laws effectively dictate the distribution structures out-of-state manufacturers must adopt to access a state’s market); Brannon P. Denning, Extraterritoriality and the Dormant Commerce Clause: A Doctrinal Post-Mortem, 73 La. L. Rev. 979 (2013).

[30] Mass. State Auto. Dealers Ass’n, Inc. v. Tesla Motors MA, Inc., 469 Mass. 675, 15 N.E.3d 1152 (2014).

[31] See Ill. Auto Dealers Ass’n v. Office of the Ill. Sec’y of State, No. 1-23-0100 (Ill. App. Ct. Aug. 23, 2024); Sarah J. Reusche, Illinois Appellate Court Affirms Direct-to-Consumer Auto Sales, Lavelle L. (Sept. 9, 2024), https://www.lavellelaw.com/illinois-appellate-court-affirms-direct-to-consumer-auto-sales.

[32] Blair et al., supra note 6.

[33] See John Smart, Rivian’s $4.6 Million Ballot Gambit to Crack Washington’s Tesla Sales Monopoly, WebProNews (Jan. 27, 2026), https://www.webpronews.com/rivians-4-6-million-ballot-gambit-to-crack-washingtons-tesla-sales-monopoly; Lisa Stiffler , The ‘Tesla exemption’ No More: Rivian and Lucid Break Through Washington State’s Dealership Wall, GeekWire (Mar. 12, 2026), https://www.geekwire.com/2026/the-tesla-exemption-no-more-rivian-and-lucid-break-through-washington-states-dealership-wall/.

[34] See Andrew J. Hawkins, Rivian Calls Ohio’s Ban on Direct Car Sales ‘Irrational in the Extreme’ in New Lawsuit, The Verge (Aug. 4, 2025), https://www.theverge.com/news/718186/rivian-ohio-lawsuit-direct-sales-ban.

[35] See Matt Wilson, Rivian Pushes for Direct Sales as Iowa Weighs Changes to Dealership Laws, Drive Tesla (Feb. 13, 2026), https://driveteslacanada.ca/news/rivian-pushes-for-direct-sales-as-iowa-weighs-changes-to-dealership-laws (discussing Iowa Senate Study Bill 3067 (proposed 2025)); Carrigan Woodson, South Carolina’s Direct-to-Consumer Car Sales Bill Ran Out of Time in 2025. Should It Be Revived in 2026?, Palmetto Promise (Aug. 5, 2025), https://palmettopromise.org/south-carolinas-direct-to-consumer-car-sales-bill-ran-out-of-time-in-2025-should-it-be-revived-in-2026.

[36] See Crane, supra note 1.

[37] See 15 U.S.C. § 1221.

[38] See Nat’l Auto. Dealers Ass’n v. FTC, No. 24-60013 (5th Cir. Jan. 27, 2025) (vacating 16 C.F.R. pt. 463 for failure to comply with advance-notice requirements under § 18 of the Federal Trade Commission Act).

[39] Thomas Randolph Beard & George S. Ford, State Automobile Franchise Laws: Public or Private Interests?, SSRN (Sept. 12, 2016).

[40] Gary Lapidus, eAutomotive: Gentleman Start Your Search Engines, Goldman Sachs (2000). The original Lapidus report is unavailable. This analysis therefore relies on Bodisch, supra note 3, which cites Lapidus and Charles H. Fine & Daniel M. G. Graff, Automotive Industry: Internet-Driven Innovation and Economic Performance, in The Economic Payoff from the Internet Revolution (Robert E. Litan & Alice M. Rivlin eds., 2001), https://archive.org/details/economicpayofffr0000unse.

[41] Bodisch, supra note 3.

[42] In 2000, the average corporate-bond yield was 7.6%; by February 2026, it had fallen to 5.3%. See Moody’s Seasoned Aaa Corporate Bond Yield (AAA), FRED, Fed. Rsrv. Bank St. Louis, https://fred.stlouisfed.org/series/AAA (last visited Feb. 25, 2026).

[43] Lapidus, supra note 40, as cited in Bodisch, supra note 3.

[44] Cox Auto Inc., supra note 4.

[45] Lapidus, supra note 40, as cited in Bodisch, supra note 3.

[46] Bodisch, supra note 3.

[47] Id.

[48] Cox Auto Inc., supra note 4.

[49] The 2000 Goldman Sachs report expressed savings as percentage shares of the average transaction price, rather than reporting individual cost inputs. The variables marked “N/A”—including floorplan rates, incentive levels, and dealership operating costs—were implicit in the original analysis but were not reported as standalone figures. Current values are provided here to show that changes since 2000 support, rather than undermine, the original ratios.

[50] See infra notes 28-34 and accompanying text.

[51] Cox Auto. Inc., New-Vehicle Inventory Returns to Pre-Tariff Levels as EV Sales Accelerate and Incentives Rise (July 2025), https://www.coxautoinc.com/insights-hub/july-2025-new-vehicle-inventory.

[52] Cox Auto Inc., supra note 4.

[53] $708 vs. $250: What Dealers Really Spend Per Car Sold, Big Time Advert. & Mktg. (2025), https://www.gowithbigtime.com/blog/C/2025/10/01/708-vs-250-What-Dealers-Really-Spend-Per-Car-Sold-177.

[54] The per-vehicle commission derives from Presidio-NCM benchmark data reporting average new-vehicle gross profit of $2,247 for full-year 2024, applied to an industry-standard 20–25% front-end commission rate ($2,247 × 0.20–0.25 = $449–$562 per vehicle). See James Hickey, Dealer Profitability Improved at End of 2024, Digital Dealer (Feb. 28, 2024), https://digitaldealer.com/news/auto-dealership-profitability-improved-at-end-of-2024/163923 (reporting Presidio-NCM benchmark data); AutoFinder, How Much Commission Does a Car Salesperson Make?, https://autofinder.com/insights/how-much-commission-does-a-car-salesman-make (reporting commission rates of 20–30% of gross profit); Mark McDonald, Car Salesman Confidential: How to Get Paid, MotorTrend (Feb. 14, 2014), https://www.motortrend.com/features/car-salesman-confidential-how-to-get-paid (reporting typical commissions of about 25% of gross profit plus backend compensation); Lawrence Hodge, How Much Car Salespeople Make in Commissions, Jalopnik (Feb. 20, 2025), https://www.jalopnik.com/1795349/how-much-car-salespeople-make-commissions (reporting commissions of roughly 20% on new-car sales).

[55] Renee Valdes, Destination Charges and Dealer Fees Explained, Autotrader (Apr. 30, 2024), https://www.autotrader.com/car-shopping/new-car-delivery-or-destination-charges-explained.

[56] Nat’l Auto. Dealers Ass’n, NADA Data: Annual Financial Profile of America’s Franchised New-Car Dealerships (2024), https://www.nada.org/nada/nada-data (reporting average dealership sales of $73.3 million with net margins of 1–2%, implying non-vehicle-acquisition operating expenses of roughly $500,000 per month or more).

[57] Cox Auto Inc., supra note 4.

[58] Id.

[59] Nathan Paulus, Average New Car Price in 2026: $48,841, MoneyGeek (Dec. 29, 2025), https://www.moneygeek.com/resources/average-price-of-a-new-car.

[60] Lapidus, supra note 40, as cited in Bodisch, supra note 3.

[61] Bodisch, supra note 3.

[62] Cox Auto Inc., supra note 4.

[63] Charles Murry & Yiyi Zhou, Consumer Search and Automobile Dealer Colocation, 66 Mgmt. Sci. 1909 (2020), https://doi.org/10.1287/mnsc.2019.3307.

[64] Florian Zettelmeyer, Fiona Scott Morton & Jorge Silva-Risso, How the Internet Lowers Prices: Evidence from Matched Survey and Automobile Transaction Data, 43 J. Mktg. Rsch. 168 (2006), https://doi.org/10.1509/jmkr.43.2.168.

[65] Cox Auto Inc., supra note 4.

[66] Lapidus, supra note 40, as cited in Bodisch, supra note 3.

[67] Bodisch, supra note 3.

[68] See Optimum Info, Dealer Financial Analysis Report—Q1 2025 (2025), https://optimuminfo.com/resources/blogs/dealer-financial-analysis-report-q1-2025; see also Harney Partners, Floor-Plan Financing for Auto Dealers: Trends, Structures & What’s Changing (2025), https://harneypartners.com/floor-plan-financing-for-auto-dealers; Bank Am., Prime Rate Information (reporting a prime rate of 6.75% effective December 2025), https://newsroom.bankofamerica.com/content/newsroom/home/prime-rate-information.html.

[69] Cox Auto. Inc., New-Vehicle Inventory Returns to Pre-Tariff Levels as EV Sales Accelerate and Incentives Rise (July 2025), https://www.coxautoinc.com/insights-hub/july-2025-new-vehicle-inventory.

[70] The $15-per-day non-interest holding cost (C??) is an estimate. No published source reports a per-vehicle, per-day figure for noninterest inventory-carrying costs. This estimate derives from aggregate dealership operating data. National Automobile Dealers Association financial profiles report average dealership payroll of about $5.39 million annually—roughly $449,000 per month—with payroll typically accounting for about 40%–50% of operating expenses. That implies total operating costs of roughly $900,000–$1.1 million per month per dealership. See Nat’l Auto. Dealers Ass’n, NADA Data: Annual Financial Profile of America’s Franchised New-Car Dealerships (2024), https://www.nada.org/nada/research-data/nada-data. To remain conservative, this brief assumes total operating costs of about $500,000 per month, only modestly above payroll and therefore likely understating the full overhead associated with maintaining dealership facilities and inventory. A typical dealership lot holds about 150–250 vehicles. See Andy Kalmowitz, Nearly 3 Million Cars Are Sitting on Dealer Lots, Jalopnik (July 12, 2024), https://www.jalopnik.com/nearly-3-million-cars-are-sitting-on-dealer-lots-1851587949 (reporting 2.89 million vehicles across roughly 17,000 lots). Allocating a conservative share of dealership operating costs to inventory maintenance—including insurance, lot upkeep, security, and depreciation exposure—yields an estimated holding cost of roughly $10–$20 per vehicle per day, assuming that about 10%–20% of dealership operating costs relate to maintaining inventory on dealership lots. This brief uses the midpoint of that range.

[71] Bd. of Governors Fed. Rsrv. Sys., Bank Prime Loan Rate [DPRIME], FRED, Fed. Rsrv. Bank St. Louis, https://fred.stlouisfed.org/series/DPRIME (last visited Feb. 25, 2026).

[72] Gérard P. Cachon & Marcelo Olivares, Drivers of Finished-Goods Inventory in the U.S. Automobile Industry, 56 Mgmt. Sci. 202 (2009), https://doi.org/10.1287/mnsc.1090.1095; Marcelo Olivares & Gérard P. Cachon, Competing Retailers and Inventory: An Empirical Investigation of General Motors’ Dealerships in Isolated U.S. Markets, 55 Mgmt. Sci. 1586 (2009), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1946447.

[73] Adam Copeland, Wendy E. Dunn & George Hall, Inventories and the Automobile Market, 42 RAND J. Econ. 121 (2011), https://www.jstor.org/stable/23046792.

[74] Lapidus, supra note 40, as cited in Bodisch, supra note 3.

[75] Nat’l Auto. Dealers Ass’n, NADA Data: Annual Financial Profile of America’s Franchised New-Car Dealerships (2024), https://www.nada.org/nada/research-data/nada-data.

[76] Nat’l Auto. Dealers Ass’n, NADA Market Beat (2025), https://www.nada.org/nada/research-data/market-beat (reporting a seasonally adjusted annual rate of approximately 16.2–16.3 million units).

[77] Mark C. Anderson, Rajiv D. Banker & Surya Janakiraman, Are Selling, General, and Administrative Costs “Sticky”?, 41 J. Acct. Rsch. 47 (2003), https://www.jstor.org/stable/3542244.

[78] Francine Lafontaine & Fiona Scott Morton, Markets: State Franchise Laws, Dealer Terminations, and the Auto Crisis, 24 J. Econ. Persp. 233 (2010), https://doi.org/10.1257/jep.24.3.233.

[79] Lapidus, supra note 40, as cited in Bodisch, supra note 3.

[80] See Cox Auto. Inc., supra note 21.

[81] Id.

[82] Ambarish Chandra, Sumeet Gulati & James Sallee, Who Loses When Prices Are Negotiated? An Analysis of the New Car Market, 65 J. Indus. Econ. 235 (2017), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2602526.

[83] Meghan R. Busse, Jorge Silva-Risso & Florian Zettelmeyer, $1,000 Cash Back: The Pass-Through of Auto Manufacturer Promotions, 96 Am. Econ. Rev. 1253 (2006), https://www.nber.org/system/files/working_papers/w10887/w10887.pdf.

[84] Andreas Grunewald et al., Auto Dealer Loan Intermediation: Consumer Behavior and Competitive Effects, Nat’l Bureau Econ. Rsch., Working Paper No. 28136 (2020), https://doi.org/10.3386/w28136.

[85] Lapidus, supra note 40, as cited in Bodisch, supra note 3.

[86] Renee Valdes, Destination Charges and Dealer Fees Explained, Autotrader (Apr. 30, 2024), https://www.autotrader.com/car-shopping/new-car-delivery-or-destination-charges-explained.

[87] Mike Monticello, The Truth About Destination Fees, Consumer Reps. (Aug. 15, 2023), https://www.consumerreports.org/cars/buying-a-car/the-truth-about-destination-fees-a1615480982; Chris Hardesty, What Are Destination Charges?, Kelley Blue Book (May 2, 2025), https://www.kbb.com/car-advice/what-are-destination-charges.

[88] See, e.g., Powers Swain Chevrolet, Will a Car Dealership Drop the Transfer Fee from Another Dealership? (July 5, 2024), https://www.pschevy.com/blog/2024/july/5/will-a-car-dealership-drop-the-transfer-fee-from-another-dealership.htm (reporting a $194.50 transfer fee for a dealer-to-dealer vehicle locate).

[89] See Carvana Co., Cost Structure Details (Nov. 2023), https://investors.carvana.com/~/media/Files/C/Carvana-IR/documents/cost-structure-details.pdf (reporting per-unit cost advantages of $1,200–$2,600 from vertically integrated inbound transportation, reconditioning, and customer delivery); see also infra § 6 (Cross-Check) (comparing Tesla’s SG&A per delivery with legacy OEMs).

[90] Bloomberg Intelligence, Retail Auto Dealers Deep Dive (June 5, 2024), https://www.madaonline.com/sites/default/files/Bloomberg_Intelligence_Retail_Auto_Dealers_June_2024.pdf.

[91] Id.

[92] See U.S. Bureau Lab. Stat., CPI Inflation Calculator, https://www.bls.gov/data/inflation_calculator.htm (last visited Feb. 25, 2026) (reporting that $26,000 in 2000 equals $50,098.06 in 2026, a cumulative increase of more than 92%).