Monopolies: Necessity or Hindrance?

Written By Joshua Smith

Introduction: The Production of Impact and the Architecture of Monopolies

The structural dominance of modern mega-corporations and geographic technology hubs presents a profound challenge to classical economic frameworks and contemporary antitrust jurisprudence. Traditional neoclassical models often struggle to differentiate between market dominance achieved through the indispensable, organic accumulation of massive infrastructure and dominance sustained through artificial market distortions and rent-seeking behavior. To accurately dissect the monopolies of Google in search, San Francisco in artificial intelligence (AI) funding, Amazon in e-commerce, and Visa and MasterCard in global payments, this analysis deploys Capacity-Based Monetary Theory (CBMT).[1]

Capacity-Based Monetary Theory postulates that money and market value are not static stores of wealth, nor are they mere fiat illusions, but rather floating-price claims on the future productive capacity of an economy.[1] This capacity is mathematically defined by an augmented Mankiw-Romer-Weil production function, where Total Impact ($I$)—the underlying collateral of a civilization—is a vector function of physical capital ($K$), human capital ($H$), the labor force ($L$), and labor-augmenting technology or efficiency ($A$).[1] The governing equation for the underlying collateral of these monopolies is expressed as:

$$I = K^\alpha H^\beta (A L)^{1-\alpha-\beta}$$

Where $\alpha$ and $\beta$ represent the elasticities of output with respect to physical and human capital, respectively.[1] In this framework, human capital ($H$) is treated not as a fungible multiplier of labor, but as a distinct, depreciating asset class that requires continuous, massive investment.[1] However, theoretical capacity is strictly constrained by the "Institutional Realization Rate" ($R_i$), a coefficient between 0 and 1 that represents the frictional costs of trust, order, the rule of law, and anti-competitive deadweight loss.[1] The realized impact of an economy is $I_{realized} = I_{theoretical} \times R_i$.[1] In a state of perfect institutional trust and competition, $R_i$ approaches 1; in a Hobbesian trap characterized by infinite transaction costs, systemic failures, or absolute monopolistic rent extraction, $R_i$ approaches 0.[1]

The central inquiry of this comprehensive report is whether the dominant market positions of Google, San Francisco, Amazon, and the Visa/MasterCard duopoly are organic derivatives of the immense physical capital ($K$) and human capital ($H$) required to operate vastly complex services—making the monopoly a structural, technological necessity—or whether these entities actively degrade the Institutional Realization Rate ($R_i$) of the broader economy through anti-competitive practices designed to lock out rivals and extract unearned rents. Through an exhaustive examination of capital expenditures, antitrust litigation, agglomeration economics, and network externalities, this report demonstrates a distinct duality. While these monopolies originated from the absolute, unavoidable necessity of scaling physical and human capital to push the technological frontier ($A$) forward, they have increasingly utilized their entrenched market positions to manipulate the institutional architecture of the market, prompting severe regulatory interventions across global jurisdictions.

San Francisco and Artificial Intelligence: The O-Ring Filter and Human Capital Agglomeration

The concentration of artificial intelligence development and venture funding in the San Francisco Bay Area defies post-pandemic expectations of decentralized, remote work. Rather than operating as a traditional corporate monopoly controlled by a single board of directors, San Francisco functions as a geographic and institutional monopoly over advanced Human Capital ($H$). The empirical data defining this agglomeration is unprecedented in modern economic history. In 2025, California-based companies captured a staggering 80% of all United States AI startup funding, representing the highest share on record.[2] Furthermore, 42% of the nation’s AI firms are clustered specifically in the Bay Area, and California accounts for over 15% of all AI-related job postings across the United States.[2, 3]

To understand why this geographic monopoly exists, one must look beyond basic networking effects and apply Michael Kremer’s O-Ring Theory of Economic Development, as integrated into the Capacity-Based Monetary Theory framework.[1, 4]

The O-Ring Theory and the Fragility of AI Production

The development of frontier generative artificial intelligence and Large Action Models (LAMs) is arguably the most complex production process currently undertaken by human civilization.[5] The O-Ring theory posits that in highly complex production processes consisting of numerous sequential tasks, a single failure or mistake by a low-skill worker in the chain destroys the value of the entire output.[4] The theory dictates that it is impossible to substitute multiple low-skill workers ($L$) for one high-skill worker ($H$) in these environments.[4]

The extreme fragility of this production chain is evidenced by the staggering failure rates of enterprise AI deployments. In 2025, American enterprises expended approximately $644 billion on AI deployments.[6] Despite this astronomical capital outlay, between 70% and 95% of those pilots failed to reach production, and a McKinsey report noted a 42% abandonment rate for enterprise AI initiatives.[6] Because the cost of failure is measured in hundreds of billions of dollars, firms must ensure that every node in their production chain is staffed by absolute top-tier talent. This imperative drives extreme assortative mating in the labor market, forcing high-skill workers to cluster together to maximize the probability of successful execution.

San Francisco acts as the ultimate O-Ring filter.[1] The city’s notoriously high cost of living, taxation, and commercial real estate operates similarly to Amotz Zahavi’s Handicap Principle within signaling theory.[1] It serves as a costly signal that reliably filters out low-capacity participants. By setting an entry cost that only agents with elite human capital ($H$) can afford, the geographic destination acts as a sorting mechanism, guaranteeing the highest talent density on the planet.[1, 2]

Physical Proximity, the Solow Residual, and Fitness Interdependence

Venture capital effectively underwrites the expected future impact of a firm that lacks current physical capital ($K$), betting almost entirely on the team’s human capital ($H$) and their ability to generate a high Solow Residual ($A$), which represents total factor productivity.[1] Despite widespread commercial office vacancies in the broader city—reaching highs of 37%—AI firms have aggressively clustered in San Francisco's SoMa, Mission Bay, and Financial District, occupying nearly 7 million square feet of premium real estate.[2, 7] Major commitments include OpenAI securing a 486,600-square-foot lease and Anthropic expanding to 650,000 square feet.[2]

This physical clustering is driven by the absolute necessity of spontaneous, high-value synergy. The ecosystem relies on physical proximity to facilitate rapid iteration, often described by insiders as "Friday debates about model architecture" that occur spontaneously in walkable neighborhoods.[2] The pipeline from elite institutions like UC Berkeley's AI Research Lab and Stanford University ensures that breakthrough academic research transitions into commercial, deployable products within months.[2] This mirrors the concept of Fitness Interdependence, where the geographic and economic survival of these engineers and founders is linked through shared equity and proximity, drastically reducing internal transaction costs and maximizing the efficiency term ($A$) in their production function.[1]

San Francisco / California AI Ecosystem Metrics (2025) Data Point Source
US AI Startup Funding Share Captured 80% [2]
National AI Firms Hosted in the Bay Area 42% [2]
Global AI Venture Dollars Captured by OpenAI & Anthropic 14% [8]
California Venture Dollars in 1H 2025 $94.5 Billion (68% of US total) [3]
AI-Related Office Space Occupied in San Francisco ~7 Million Sq. Ft. [2]
Enterprise AI Spend (US) $644 Billion [6]
Enterprise AI Pilot Abandonment Rate 42% [6]

In this context, San Francisco's monopoly on AI funding and talent is not an anti-competitive market failure artificially orchestrated by a cartel. It is a structural necessity derived directly from the inherent complexity of the technology. The sheer difficulty of training and aligning multi-modal AI models requires a density of human capital ($H$) and an efficiency of interaction ($A$) that cannot be replicated in a distributed, remote-work paradigm or dispersed across secondary cities. The monopoly is a required condition to operate at the bleeding edge of complicated computational sciences.

However, this geographic monopoly is not without systemic risks. The extreme concentration of capital creates a highly speculative environment, drawing comparisons to historical asset bubbles. If the massive capital expenditures do not yield proportional gains in economic productivity, the resulting collapse in venture valuations could trigger a severe regional economic contraction.[9] Yet, regarding the specific question of whether this monopoly hinders progress through anti-competitive practices, the evidence suggests the opposite: the agglomeration is the very engine enabling the rapid advancement of the technological frontier.

Google Search: The Physical Capital Behemoth and the Behavioral Manipulation of Defaults

While San Francisco exemplifies an organic monopoly of human capital ($H$), Google’s overwhelming dominance in general internet search represents a monopoly born from unprecedented physical capital ($K$) requirements, which has subsequently been fortified and maintained through the deliberate, anti-competitive manipulation of the Institutional Realization Rate ($R_i$).

Google maintains an estimated 90% share of the global search engine market in 2026, processing an astounding 9.5 million searches every minute, drawing from a search index that exceeds 100,000,000 gigabytes.[10, 11] The foundational argument for Google as a "natural monopoly" rests securely on the sheer scale of the infrastructure required to continuously crawl, index, and rank the exponentially expanding internet.

The Immense Physical Capital ($K$) Requirement of Web Indexing

The financial cost of indexing the web and maintaining the requisite hyperscale data centers constitutes a near-insurmountable technical and economic barrier to entry.[12] To support its core search functions, alongside its aggressive expansion into generative AI infrastructure, Google's parent company Alphabet reported capital expenditures of \$52.5 billion in 2024 and \$91.4 billion in 2025.[13] In 2026, executives announced plans to elevate CapEx to upwards of \$185 billion globally, primarily directed toward advanced servers, networking equipment, and the acceleration of data center construction to meet compounding computational demands.[13, 14, 15]

For a nascent competitor to replicate this index from scratch, the capital requirements are entirely prohibitive. Even Apple, a corporation with vast financial resources, estimated it would cost an additional \$6 billion annually in ongoing operational costs just to match Google's search and indexing capabilities, completely independent of the initial capital outlay required to build the infrastructure.[16] Thus, under the strict Capacity-Based Monetary Theory framework, the initial monopolization of the search market is a direct, unavoidable result of the immense physical capital ($K^\alpha$) necessary to produce the required efficiency ($A$) and utility for the end user. The search engine is a vast mechanical Turk—a reinforcement learning engine where extreme scale creates a virtuous cycle of data refinement that no sub-scale competitor can match.[16]

The DOJ Case: Transitioning from Natural Monopoly to Anti-Competitive Exclusion

If Google's market dominance rested purely on its superior physical infrastructure ($K$) and algorithmic efficiency ($A$), it would not necessitate the expenditure of tens of billions of dollars annually to manipulate user behavior. However, the United States Department of Justice (DOJ) successfully argued, and a federal court affirmed, that Google illegally maintained its monopoly through a vast network of exclusionary default contracts.[17, 18]

The most prominent of these is Google's Information Services Agreement (ISA) with Apple, under which Google pays an estimated \$18 billion to \$20 billion annually to remain the undisputed default search engine on all iOS devices.[19, 20] This single partnership drives nearly 50% of Google's search traffic.[19]

By paying \$20 billion annually to secure default placement, Google is effectively creating a Hobbesian transaction cost for its competitors.[1] This exorbitant payment does not improve the production function; it adds no physical capital, no human capital, and no algorithmic efficiency to the search index. Rather, it artificially suppresses the Institutional Realization Rate ($R_i$) of rivals like Microsoft's Bing, DuckDuckGo, or emerging AI-native search engines. It buys the behavioral architecture of the internet.

The mechanics of this behavioral barrier were rigorously explored in a 2025 National Bureau of Economic Research (NBER) randomized controlled trial involving 2,354 desktop users.[21] The study sought to measure the precise factors explaining Google's dominance. The findings were revelatory: high switching costs (the actual effort required to change search engines) are not the primary barrier. Instead, user inattention and the overwhelming power of defaults dictate market share.[21] When users in the study were forced to make an "active choice" regarding their search engine, a significant portion deviated from Google, proving that the default status, rather than pure product superiority, sustains the monopoly.[21] Therefore, Google's modern monopoly is no longer purely a function of its superior index ($K$), but of its financial ability to buy the behavioral pathways of consumers, actively hindering the progress of competitors.

Google Search Monopoly Metrics (2024-2026) Data Point Source
Global Search Market Share ~90% [11]
Search Volume 9.5 Million per minute [10]
Estimated Size of Search Index > 100,000,000 GB [10]
Alphabet Capital Expenditures (2025) \$91.4 Billion [13]
Alphabet Capital Expenditures (2026 Est.) \$185 Billion [14]
Annual Default Payment to Apple \$18 Billion - \$20 Billion [20]
Search Revenue (2025) \$63.1 Billion [11]

Judicial Remedies and the Future of the Search Algorithm

In August 2024, U.S. District Court Judge Amit Mehta ruled decisively that Google violated Section 2 of the Sherman Act by maintaining monopolies in general search services and general text advertising.[17] The subsequent remedies ordered in late 2025 stopped short of aggressively breaking up the company—such as forcing the divestiture of the Chrome browser or the Android mobile operating system—recognizing that such draconian structural remedies could severely disrupt the integrated ecosystem and harm consumers.[22, 23]

Instead, the court pursued a remedy perfectly aligned with the CBMT framework. The court ordered Google to share targeted portions of its underlying search index and user-interaction data with competitors for a period of five years, while simultaneously prohibiting future exclusive default contracts across devices and browsers.[17, 18] By forcing Google to share its index data, the judiciary is artificially transferring a portion of Google's accumulated physical capital ($K$) and historical human capital ($H$) to rivals. This intervention attempts to lower the insurmountable barrier to entry and restore a competitive Institutional Realization Rate ($R_i$) to the broader digital ecosystem.[24, 25]

However, the rapid integration of artificial intelligence is fundamentally altering this landscape before the remedies can fully take effect. Google's aggressive rollout of "AI Overviews" at the top of search results has already caused a massive paradigm shift. Industry data indicates that AI Overviews caused a 68% decline in click-through rates (CTR) to third-party websites for certain query categories between mid-2024 and late 2025.[26] This phenomenon, dubbed "The Great Decoupling," results in 60% of Google searches ending without a single click to an external website.[27] Google is leveraging its illegally maintained monopoly position in search to rapidly gain a foothold in the nascent market for AI-powered answer engines.[28]

Ultimately, Google's search monopoly began as an absolute necessity of scale—no entity could map the internet without hundreds of billions in capital. Yet, it steadily evolved into a legally recognized anti-competitive structure reliant on exclusionary contracts, proving that while the infrastructure is necessary, the monopolistic business practices actively hinder digital progress.

Amazon: The Logistics Flywheel, Algorithmic Pricing, and the Extraction of Seller Rents

Amazon represents the most complex and multifaceted intersection of inherent capital necessity and anti-competitive rent-seeking in the modern global economy. Capturing a verified 37.6% of the United States e-commerce market in 2024, generating an estimated $447.4 billion in U.S. online retail revenue [29], Amazon operates a highly integrated business model often described as a "flywheel".[30] This flywheel connects the extraordinarily high-margin profits of its cloud computing division (Amazon Web Services, or AWS) and its burgeoning advertising services network with the notoriously low-margin, high-volume operations of retail and physical logistics.[30]

The Absolute Necessity of $K$: The Fulfillment Network

Amazon’s dominance in e-commerce is fundamentally underpinned by a physical logistics and data center infrastructure that is virtually impossible for any new entrant, or even established legacy retailers, to replicate. In 2025, Amazon's total domestic investment in the United States exceeded \$340 billion, an amount encompassing physical infrastructure expansion and employee compensation.[31] Looking forward, the company has committed to an unprecedented, staggering \$200 billion in global capital expenditures for the fiscal year 2026, heavily driven by the deployment of AI infrastructure, custom silicon (such as Trainium2 chips), and the continued expansion of AWS.[30, 32, 33]

The physical reality of Amazon's Fulfillment by Amazon (FBA) network—comprising thousands of massive fulfillment centers, regional sortation facilities, advanced autonomous robotics, and an immense last-mile delivery fleet—requires an astronomical input of physical capital ($K^\alpha$).[33, 34] In the second quarter of 2025 alone, Amazon's fulfillment costs reached \$25.9 billion, an operating expense that covers the labor, leasing, and depreciation required to maintain this vast network.[35]

This massive physical capital accumulation allows Amazon to achieve an unparalleled efficiency term ($A$) in the Mankiw-Romer-Weil equation. The result is an infrastructure that generates immense consumer surplus. A 2024 independent study demonstrated that prices in Amazon's store were, on average, 14% lower than all other major U.S. retailers across all product categories.[36] Furthermore, the network delivers unprecedented shipping speeds, fundamentally altering consumer expectations globally. Under the strict CBMT specification, Amazon is optimizing the production function for maximum retail impact. The sheer complexity of moving millions of physical goods globally within 48 hours absolutely necessitates this monopolistic scale; fragmented, sub-scale competitors simply cannot match the unit economics of Amazon's logistics network.

Project Nessie, FBA Tying, and Algorithmic Collusion: The FTC’s Allegations

However, the immense benefits provided to the consumer do not negate the anti-competitive mechanisms utilized to sustain the ecosystem. In the fall of 2023, the Federal Trade Commission (FTC), alongside 19 state attorneys general, filed a sweeping, landmark antitrust lawsuit against Amazon, alleging that the company operates as an illegal monopoly that utilizes interlocking anticompetitive strategies to stifle innovation, overcharge sellers, and ultimately harm consumers.[37, 38] The core of the FTC's argument is that Amazon deliberately degrades the Institutional Realization Rate ($R_i$) for independent sellers and rival platforms, raising transaction costs across the entire internet economy.

The FTC allegations center on three primary mechanisms of market manipulation:

  1. The First Anti-Discounting Algorithm: Amazon deploys an expansive surveillance network to monitor the prices of similar goods across the entire internet. If Amazon detects that a third-party seller is offering a product at a lower price on a competing website (e.g., Walmart.com or their own direct-to-consumer site), Amazon algorithmically punishes the seller by removing them from the "Buy Box".[37, 39] Because approximately 98% of all Amazon sales occur through the Buy Box, losing access is economically devastating.[37] This forces sellers to use their Amazon price as the absolute price floor across the internet, artificially inflating prices across all competing retail platforms.[39]
  2. Tying Prime Eligibility to Fulfillment by Amazon (FBA): The FTC alleges that Amazon unfairly restricts competition among logistics providers by premising a seller's access to the coveted "Prime" badge on their mandatory use of Amazon's exclusive fulfillment service, FBA.[40, 41] Because independent merchants generally lack the capital to utilize multiple disparate logistics services simultaneously, tying Prime sales to FBA exploits Amazon's market dominance to lock out competing shipping networks and forces sellers into Amazon's fee structure.[40]
  3. Project Nessie: Perhaps the most sophisticated allegation involves a secret algorithmic pricing tool known internally as "Project Nessie," which Amazon utilized between 2014 and 2019.[37, 42] Nessie was an algorithm designed to predict whether competing retailers would match Amazon's price increases. If the algorithm determined a match was highly probable, it would intentionally raise Amazon's prices, effectively inducing competitors to follow suit.[37, 42] This resulted in coordinated overcharges for shoppers both on and off the Amazon platform.

Project Nessie highlights a profound and alarming evolution in modern monopolies: the use of artificial intelligence and adaptive learning algorithms to achieve implicit, tacit collusion without any traditional, illegal communication between executives. Academic research from Carnegie Mellon University indicates that when advanced AI algorithms (utilizing reinforcement learning) compete against simple rule-based algorithms (like automated "tit-for-tat" price matching), the AI quickly learns to strategically raise prices. The AI understands that its competitors will blindly match the increase, thereby boosting profits for all sellers at the direct and severe expense of consumer surplus, creating substantial deadweight loss.[43] The FTC successfully argued that this constitutes an unfair method of competition under Section 5 of the FTC Act, marking the first time in over 40 years that a standalone Section 5 claim survived a motion to dismiss in federal court.[37]

The Escalation of Seller Fees: A Deadweight Loss Analysis

The most direct, empirical evidence of Amazon's unchecked monopoly power—defined economically as the ability to raise prices above competitive levels without suffering a commensurate loss in market share—is found in its increasingly aggressive treatment of third-party sellers. Over the last decade, Amazon's extraction of revenue from independent merchants has escalated dramatically.

Reports from research groups indicate that Amazon's total "take-rate"—the percentage of a seller's revenue that Amazon retains through mandatory referral fees, FBA fulfillment charges, and virtually required advertising spend—has exploded from an average of 19% in 2014 to roughly 45% in 2023 and 2024.[39, 44] In 2024 alone, seller fees generated over \$150 billion in revenue for Amazon, a figure so substantial it would qualify as a Fortune 25 company independently.[44] Crucially, these seller fees now account for 29% of Amazon's non-AWS revenue, a 53% proportional increase in just five years.[44]

These fees act as an inescapable, monopolistic tax on the entire e-commerce ecosystem. Sellers are increasingly forced into what industry analysts term the "Hidden Cost Trap," navigating constant, opaque increases in dimensional weight pricing, inbound placement fees, low-inventory surcharges, and aggressive aged-inventory penalties.[45, 46, 47] Because sellers cannot afford to leave the platform that controls nearly 40% of the market, they are forced to absorb these costs until they inevitably pass them on to consumers through higher retail prices.

Amazon Financial & Market Metrics (2024-2026) Data Point Source
U.S. E-commerce Market Share (2024) 37.6% [29]
Total U.S. E-commerce Market Size (2024) $1.19 Trillion [29]
Amazon Global CapEx (2026 Estimate) $200 Billion [30]
Third-Party Seller Estimated Take-Rate ~45% of seller revenue [39, 44]
Seller Fee Revenue (2024) > $150 Billion [44]
Q2 2025 Fulfillment Costs $25.9 Billion [35]

Interestingly, empirical economic research following the announcement of the FTC's antitrust allegations suggests that the mere threat of severe regulatory intervention altered Amazon's behavior. A study analyzing product pricing and fee structures found that Amazon reduced its FBA fees by approximately \$0.27 to \$0.29 per product within six months of the FTC's allegations.[48] This slight reduction, driven purely by regulatory pressure to uphold public trust and appease regulators, is estimated to save FBA sellers between \$0.85 billion and \$0.92 billion annually, actively reducing the deadweight loss in the market.[48]

In conclusion, Amazon’s fulfillment and cloud networks absolutely necessitate monopolistic scale ($K$) to function at their current, miraculous efficiency. A fragmented logistics market could never provide two-day nationwide shipping at current cost structures. However, the aggressive algorithmic policing of off-platform prices, the tying of essential services, and the relentless, unchecked extraction of seller fees demonstrate unequivocally that Amazon utilizes this necessary infrastructure to impose severe transaction costs on the broader retail market. The monopoly is required to operate the service, but its current business practices actively hinder the economic progress of independent merchants and artificially inflate prices across the digital economy.

Visa and MasterCard: Network Externalities and the Two-Sided Market Architecture

The global payments duopoly of Visa and MasterCard presents a fundamentally different structural and architectural model from the primary producers of goods (Amazon) or information indexes (Google). Visa and MasterCard do not manufacture physical products, nor do they hold consumer deposits or issue credit directly. Rather, they are the vital "software" of the global economy. In the context of Capacity-Based Monetary Theory, they directly provide the institutional trust and verification ($R_i$) required to escape the Hobbesian trap of counterparty risk in instantaneous, cross-border commerce.[1]

Visa and MasterCard operate classic "two-sided markets," a complex economic structure where the platform must simultaneously balance and incentivize the participation of two distinct user groups: merchants (who must be willing to accept the cards) and consumers/issuing banks (who must be incentivized to carry and use the cards).[49] The scale of these networks is staggering and deeply entrenched. In fiscal year 2025, Visa processed an astonishing 257.5 billion transactions, facilitating \$16.7 trillion in total payments volume across 4.9 billion active payment credentials globally.[50] MasterCard processes similarly massive volumes, driving net revenues of \$28.2 billion in 2024.[51]

The Enormous Capital Requirements of Global Trust

Maintaining a ubiquitous payment network that operates seamlessly, instantly, and securely across hundreds of different sovereign borders and fiat currencies requires continuous, massive investment in both labor-augmenting technology ($A$) and highly secure physical infrastructure ($K$). Visa operates four primary, global hyperscale data centers that feature extreme redundancy, network connectivity, power backup, and advanced cooling systems designed to provide absolute, continuous availability of financial systems.[52]

Furthermore, the cybersecurity requirements to protect this volume of capital transfer are monumental. Visa has invested over $3 billion specifically in artificial intelligence and data infrastructure over the past decade to enhance predictive fraud detection and network security.[52] This infrastructure is not an optional luxury; it is an absolute necessity. The cost of credit card fraud to the financial system is immense. A comprehensive 2025 study determined that for every single dollar of face-value fraud loss incurred, the actual, total cost to U.S. lenders is 5.4 times higher due to downstream operational impacts, risk management, and labor-intensive recovery processes.[53]

By socializing the exorbitant costs of cybersecurity, network tokenization, and real-time ledger processing across tens of trillions of dollars in transaction volume, Visa and MasterCard achieve economies of scale that no individual regional bank, credit union, or independent merchant could ever hope to replicate.[52, 54] The monopoly (or strict duopoly) is therefore a natural, inevitable byproduct of extreme network externalities—the payment system becomes exponentially more valuable and secure to both merchants and consumers as more global participants join.[54]

The Interchange Fee Dispute: System Maintenance vs. Monopolistic Rent Extraction

Despite the undeniable, foundational value of the network infrastructure, the pricing structure dictated by the duopoly—specifically the "interchange fee" (colloquially referred to as a "swipe fee")—has been the subject of decades of bitter, intense antitrust litigation and legislative battles.

Interchange fees are not paid directly to Visa or MasterCard. Rather, they are paid by the merchant's acquiring bank to the consumer's card-issuing bank, though the network operators (Visa/MasterCard) centrally set the rates and rules governing these transfers.[55] In 2025, Visa's credit interchange rates range generally from 1.30% to 2.60% per transaction, while MasterCard's rates range from 1.45% to 2.90%, depending heavily on the type of card used (e.g., standard vs. premium rewards cards) and the merchant category.[56, 57] In 2023, these fees cost U.S. merchants and, by extension, consumers upwards of \$133.75 billion, a figure that rose to an estimated, record-breaking \$148.5 billion in 2024.[58, 59]

Merchants and retail advocacy groups argue vehemently that these fees operate as monopolistic, inescapable taxes enforced through anti-competitive contractual restraints.[59] Historically, networks enforced "honor-all-cards" rules and network exclusivity agreements, obligating merchants who accepted basic Visa cards to also accept ultra-premium rewards cards that carry significantly higher interchange fees, while simultaneously preventing merchants from steering consumers to cheaper payment methods or applying surcharges to offset the specific costs of premium cards.[59, 60]

Conversely, the payment networks and issuing banks argue that interchange fees are simply the cost of doing business, funding the vital fraud prevention architecture, guaranteeing immediate payment to the merchant, and subsidizing highly popular consumer rewards programs (cashback, airline miles) that ultimately drive increased retail spending and macroeconomic velocity.[59, 60]

The 2024/2025 Settlement and the Danger of Legislative Price Controls

The tension between necessary system funding and monopolistic rent extraction culminated in a landmark legal resolution. In March 2024, after nearly twenty years of grueling antitrust litigation, Visa and MasterCard agreed to a historic settlement with U.S. merchants, over 90% of which are small businesses.[61, 62]

The proposed settlement provides estimated relief of over $30 billion to merchants.[62] Crucially, it caps standard U.S. consumer credit card interchange rates at 1.25% for a period of eight years, and mandates a reduction of the published effective rate by 10 basis points for a period of five years, providing unprecedented cost certainty.[63, 64] More importantly from an antitrust perspective, the settlement fundamentally alters network rules, providing merchants with much greater point-of-sale flexibility. Merchants will now have the optionality to surcharge specific card brands or categories, and the ability to actively steer customers to lower-cost preferred payment methods.[61, 63, 65]

When evaluating whether the Visa/MasterCard duopoly fundamentally hinders progress, one must examine the historical impact of direct government price controls on complex payment networks. The most prominent example is the 2010 Durbin Amendment to the Dodd-Frank Act, which sought to alleviate merchant burdens by capping debit card interchange fees for banks with over \$10 billion in assets at a maximum of 21 cents plus 0.05% of the transaction value.[66]

While this legislation saved merchants an estimated \$8.5 billion in its first year, it yielded severe, highly regressive unintended consequences. Because the regulation forced a strict issuer-cost-based model and completely ignored the delicate, cross-subsidizing nature of a two-sided market, the card networks responded rationally to preserve revenue: they raised the fees on small-value purchases to the maximum allowable cap level.[66] Consequently, small merchants processing low-ticket items (e.g., coffee shops) suddenly faced higher relative costs, while large big-box retailers benefited massively.[67] Furthermore, research indicated that the Durbin amendment created a regressive wealth transfer, where low-income, cash-using households effectively subsidized the system for affluent card-using households, with each cash-using household transferring approximately \$149 annually to card-accepting merchants.[68]

Payment Network Metrics (2024-2025) Data Point Source
Visa Total Payment Volume (FY 2025) \$16.7 Trillion [50]
Visa Processed Transactions (FY 2025) 257.5 Billion [50]
Total Swipe Fees Paid by U.S. Merchants (2024) \$148.5 Billion [58]
Visa Estimated Interchange Range (2025) 1.30% - 2.60% [56]
MasterCard Estimated Interchange Range (2025) 1.45% - 2.90% [56]
Merchant Settlement Relief > \$30 Billion [62]
Settlement Rate Cap (Standard Consumer) 1.25% for 8 years [63]

Thus, under the CBMT framework, while Visa and MasterCard extract a significant economic premium for operating the trust layer of the economy, forcibly altering their intricate pricing structure via heavy-handed legislative price controls (such as the heavily debated Credit Card Competition Act) often violently distorts the Institutional Realization Rate ($R_i$) rather than organically optimizing it.[66, 69] The 2024/2025 class-action settlement, which focuses intensely on enhancing merchant choice, transparency, and competitive steering capabilities rather than enforcing strict legislative price ceilings, represents a far more market-aligned, sophisticated approach to checking the duopoly's formidable market power.[61, 64]

Synthesis: Capacity-Based Monetary Theory and the Duality of Modern Monopolies

Applying Capacity-Based Monetary Theory to these four distinct, massive entities reveals a unifying, profound ontological truth about modern market dominance: absolute scale is no longer an optional business strategy; it is an unavoidable technological prerequisite.

  1. The Absolute Necessity of Scale ($K$ and $H$): None of these monopolies could operate their core, foundational services without their current, unprecedented scale. Google cannot index over 100 million gigabytes of the rapidly expanding web and serve 9.5 million queries a minute without committing to \$185 billion in ongoing Capital Expenditures. San Francisco cannot incubate fragile, frontier Large Action Models without the extreme concentration of elite human capital facilitated by its high-cost O-Ring filter. Amazon cannot reliably fulfill millions of retail orders globally within 48 hours without its massive, heavily integrated warehousing and robotics infrastructure. Visa cannot secure \$16.7 trillion in global commerce without socializing the immense cost of AI-driven fraud detection across hundreds of billions of transactions. In the strict terms of the Mankiw-Romer-Weil equation, these entities have logically maximized physical capital ($K^\alpha$) and human capital ($H^\beta$) to push the technological efficiency frontier ($A$) forward for human civilization.[1] The monopolies are structurally necessary to provide the services at the current level of expected utility.

  2. The Willful Degradation of Institutions ($R_i$): However, the transition from a benign "necessary monopoly" to a parasitic "anti-competitive monopoly" occurs reliably when the entity realizes that maintaining its absolute dominance through continuous physical and human capital investment is ultimately more expensive, or less certain, than artificially manipulating the behavioral, legal, and contractual architecture of the market.

This is the precise crux of the intense global antitrust scrutiny facing these firms. Google's \$20 billion annual payment to Apple is a synthetic, behavioral barrier to entry, not an improvement in search technology or consumer utility. Amazon's deployment of Project Nessie and its relentless extraction of up to 45% of independent seller revenue represent the weaponization of its indispensable platform to extract unearned rents, driving up consumer prices implicitly across the entire e-commerce ecosystem. Visa and MasterCard's historical reliance on "honor-all-cards" rules forced merchants to accept exorbitant fees on premium rewards cards, explicitly restricting free-market steering and price discovery.

These deliberate corporate actions artificially and maliciously lower the Institutional Realization Rate ($R_i$) for competitors, merchants, and emerging innovators. They introduce Hobbesian transaction costs back into an economic system that the digital platforms originally promised to streamline and democratize.

Conclusion

The structural monopolies of Google in search, San Francisco in AI funding, Amazon in e-commerce, and Visa and MasterCard in global payments are derived fundamentally from inherent technological complexity and the massive, unprecedented capital required to build and operate modern digital and physical infrastructure. The barriers to entry—whether the cost of building a global hyperscale server network, the geographic clustering of elite AI PhDs to prevent production chain failure, the deployment of a continent-spanning logistics fleet, or the maintenance of a highly secure, instantaneous financial routing network—are organic, logical, and structurally necessary. Sub-scale competitors simply cannot execute these tasks efficiently.

However, recognizing the structural necessity of their massive scale does not absolve these corporations of anti-competitive behavior. The empirical evidence, validated by federal courts and antitrust regulators globally, overwhelmingly indicates that once these entities achieved their natural, capital-driven dominance, they actively deployed exclusionary contracts, algorithmic price-fixing, and unavoidable ecosystem taxes to insulate themselves from future competition and extract outsized rents from dependent participants.

Economic progress is not hindered by the mere existence of their massive infrastructure; rather, it is choked by the frictional, anti-competitive constraints these monopolies place on the merchants, consumers, and innovators who have no choice but to interface with it. The most economically sound and effective regulatory responses—such as the DOJ's mandate for Google to share its underlying search data, or the Visa/MasterCard class-action settlement granting merchants the right to steer payments—are those that carefully preserve the immense efficiency and utility of the centralized infrastructure (maximizing $K$ and $A$), while surgically dismantling the artificial, contractual barriers that degrade the broader economic ecosystem's Institutional Realization Rate ($R_i$). By managing capacity and ensuring open access rather than simply punishing scale, policymakers can ensure that these necessary monopolies continue to drive the future impact of civilization without cannibalizing the very free markets that birthed them.

Joshua Smith
Next

Japan's Economic Growth Strategy