Executive summary
Conglomeration is often sold as efficiency. Bigger firms can standardize operations, finance long-term investment, and move goods through the economy at low cost. Scale can also create fragility.
When essential services and bottlenecks run through a small number of corporations, failures stop being private events and become public problems. Outages cascade and shortages spread. A recall or cyber incident can disrupt households, hospitals, and state agencies at the same time.
The core argument is simple. Some corporations now function like critical infrastructure without infrastructure-style oversight. The policy question is not whether large firms are inherently bad. The question is whether the economy should accept single points of failure and concentrated power without the safeguards that apply to other forms of infrastructure.
This piece explains what concentration looks like in practical terms, why optimization often turns into brittleness, how conglomerates can become too essential to fail even outside finance, and what a pragmatic agenda looks like when competition policy is blended with resilience and governance tools.
Series: System Reform (Part 6 of 7)
Strategic position: Bridges financial regulation and antitrust. Applies systemic-risk thinking beyond banks.
Visuals
Visual 1: HHI thresholds (DOJ and FTC)
| HHI range | How agencies describe it | Why it matters |
|---|---|---|
| Below 1,000 | Unconcentrated | Structure alone usually does not trigger a structural presumption |
| 1,000 to 1,800 | Moderately concentrated | Concentration is a warning signal, but context matters |
| Above 1,800 | Highly concentrated | Highly concentrated markets raise structural risk and can trigger stronger presumptions in merger review |
Caption: DOJ Antitrust Division HHI thresholds and the 2023 Merger Guidelines.
I. Introduction: The illusion of choice
Walk down a grocery aisle and it looks like abundance. Hundreds of labels appear to compete. Multiple price points coexist. Premium brands sit next to store brands.
Now look behind the shelf. The same parent companies appear again and again. Many “competitors” share suppliers, logistics networks, and advertising platforms. They buy the same inputs and distribute through the same channels. In some categories, a small set of firms decides what gets produced, how it gets shipped, and what happens when something goes wrong.
This is not a niche concern. Concentration and conglomeration shape everyday life, influencing food and medicine, cloud computing, transportation, media, and finance. These systems are not only markets. They are networks, and when nodes become too dominant, networks can behave like infrastructure.
Infrastructure has two defining traits. If it fails, the damage is widespread. If it is controlled by a small number of actors, the public has limited options. A power grid is infrastructure. A port is infrastructure. In modern economies, a small number of private corporations can also become infrastructure.
This creates a governance gap. The economy relies on concentrated private systems, but policy still treats most of them as ordinary firms. That mismatch is the heart of the problem.
II. What concentration is, and how to measure it
Concentration is not one thing. Some markets have a few big firms and intense rivalry. Others have a few big firms and a quiet truce. Some markets have many brands but a concentrated set of owners. Some markets look competitive at the retail level but are concentrated in upstream bottlenecks such as processing, logistics, cloud services, or app distribution.
A quick primer on HHI
The most common concentration measure in antitrust is the Herfindahl-Hirschman Index (HHI). It is built from market shares and squares them, which makes dominant players matter more. HHI is not a moral judgment. It is a diagnostic.
The DOJ and FTC use HHI thresholds as one input in merger review, and the threshold language matters because it turns “concentration” into something measurable and reviewable. A key limitation is that HHI requires a clear definition of a market. In real life, market boundaries are contested, and firms also compete across categories and ecosystems. Even with those caveats, concentration metrics are useful because they reveal when markets stop being self-correcting.
Concentration is often a bottleneck problem
Many of the most consequential concentration risks are not the ones consumers notice. Consumers see retail choice. Systems depend on chokepoints.
A chokepoint can be a small number of meat processors, a small number of infant formula plants, a small number of cloud providers, a small number of payment networks, or a small number of shipping and freight networks. When a chokepoint fails, alternatives are limited and ramping capacity takes time. Regulators become crisis managers, and governments face pressure to intervene. This is how private scale turns into systemic risk.
III. The efficiency paradox: how optimization creates fragility
The case for scale is real. Large firms can reduce unit costs, stabilize supply, and fund R&D. They can coordinate complex systems and spread fixed costs over millions of customers.
The paradox is that the same forces that make systems efficient often remove the slack that makes them resilient.
Redundancy looks like waste until you need it
Redundancy can take the form of inventory, spare capacity, multiple suppliers, or extra routing options. In many sectors, cost pressure pushes the system to remove redundancy. Firms shift to just-in-time inventory, consolidate suppliers, standardize components, and centralize production.
These moves can be rational at the firm level. They can also create a system that fails in correlated ways. When redundancy is removed across an entire sector, the system becomes brittle. Small shocks propagate, recovery takes longer, and failures become more politically salient.
Single points of failure are not theoretical
The pattern is visible across recent crises. A recall, a cyber incident, a natural disaster, or a factory shutdown can cause nationwide disruption when production is concentrated. This is not always “the fault” of a firm. It is a feature of how the system is structured. Policy should care about structure, not only about misconduct.
IV. When competition becomes theoretical
Market power can be subtle. It is not only about raising prices. It can show up in the ability to set terms for suppliers and sellers, to bundle products and steer demand, to cross-subsidize with profits from a separate business line, to acquire potential rivals early, and to shape standards, rules, and distribution.
Conglomerates can hide power in plain sight
A conglomerate can be present across multiple markets at once, and each market might look competitive on its own. The conglomerate can still have leverage because it can bundle services, use data across products, threaten to retaliate across multiple lines of business, and finance a price war in one market using profits from another. This is one reason “consumer choice” often fails as a check on corporate power. Choice is not only about what is on the shelf. It is about what is feasible for competitors to build.
Vertical integration can become a governance problem
Vertical integration can reduce transaction costs. It can also create conflicts of interest when a firm controls both a platform and a competing product.
The issue is not size alone. The issue is a structure where the platform sets rules, the platform sees competitor data, the platform can self-preference, and the platform can change access terms overnight. In those settings, market competition becomes theoretical. Entry becomes possible in principle and difficult in practice.
V. Too big to fail, outside finance
“Too big to fail” is often treated as a banking story. The underlying logic is broader.
A firm can become too essential to fail when substitutes are limited, when the time to replace the service is long, when the service is embedded in other systems, and when failure would impose immediate, widespread costs.
Critical infrastructure is increasingly private and concentrated
A modern economy relies on private infrastructure layers, including cloud computing and data hosting, payment networks, logistics and delivery networks, communications platforms, and specialized manufacturing nodes. These systems can have healthy competition at the margin while still being concentrated enough to create systemic risk.
A major outage can disrupt firms, governments, hospitals, and households simultaneously. This shifts the policy problem. The question becomes what the public should expect from systems that the public cannot easily replace.
Moral hazard is not limited to banks
When firms know they are essential, incentives change. Risk-taking can increase, cost-cutting can become more aggressive, and investments in resilience can be deferred. This does not require bad actors. It can emerge from standard corporate incentives such as quarterly performance pressure, investor expectations, and executive compensation structures.
If public actors will absorb the cost of failure, private actors will not fully price that risk. This is the classic moral hazard logic, and it applies to any systemically important firm, not only a bank.
VI. The incentive problem: growth, capital markets, and empire building
Conglomeration is not only a policy failure. It is often an incentive outcome.
Growth is rewarded, even when it increases fragility
Public markets reward growth and often punish stability. In mature markets, organic growth slows and firms look for new revenue streams. They expand into adjacent sectors, buy competitors, buy suppliers, and buy distribution.
Conglomeration can become a strategy for diversifying revenue, smoothing earnings, increasing bargaining power, and controlling bottlenecks. These incentives can align with shareholder value. They can diverge from system resilience.
Complexity can create opacity
Complex organizations are harder to monitor. They can also be harder to regulate because regulators are typically organized by sector, while conglomerates span sectors. This creates blind spots.
A problem in one division can affect the balance sheet, credit, or operational integrity of the whole enterprise, and the links are often hard to see until a stress event occurs. Complexity can also reduce accountability. When responsibility is spread across units, it becomes easier to blame “the system” and harder to identify actionable fixes.
VII. The new feudalism: private power with public consequences
The phrase “corporations as governments” is often used rhetorically. There is a concrete version that matters.
A corporation becomes quasi-governmental when it sets rules for participants, enforces those rules, controls access to essential distribution, and operates with limited external oversight. Examples vary by sector. The mechanisms are consistent.
This is not primarily about ideology. It is about governance. When private actors set rules with public consequences, democratic systems struggle to keep up.
VIII. What the evidence tends to show
This article is not a single empirical claim. It is a synthesis of mechanisms. Even so, the evidence base matters because concentration debates often degrade into assertions.
Across the literature, several patterns appear frequently. Concentration can increase markups and pricing power in some sectors. Concentration can reduce entry and innovation, especially when incumbents can acquire potential challengers. Concentration can weaken labor bargaining power in local labor markets. Concentration can create systemic risk by concentrating production, logistics, and operational dependencies.
These are not universal. They are context-dependent. The key point is that concentration changes the distribution of risk. It moves risk from firms to households and the public sector.
IX. A pragmatic policy agenda
A practical agenda should be built around a simple principle. If a private firm functions like infrastructure, policy should require infrastructure-like safeguards.
This does not imply government ownership. It implies rules that reduce single points of failure, constrain conflicts of interest, and improve accountability.
1. Stricter merger standards, focused on structure
Modern merger review should weigh more than short-run price effects. A better standard includes market structure and entry barriers, labor market effects, innovation impacts, and systemic risk and resilience.
The goal is not to block every large merger. The goal is to prevent mergers that create chokepoints, increase correlated failure risk, and make future enforcement impossible.
2. Structural separation when platforms compete on their own platform
When a firm both operates a platform and competes on it, conflicts of interest are structural. Policy options include separation of platform and commerce units, non-discrimination requirements, and strong transparency and audit regimes. The preferred approach depends on the sector. The key principle is consistent. A referee cannot also be a player without very strong constraints.
3. Utility-style obligations for systemically important services
Some services merit obligations similar to utilities. These can include non-discrimination, reliability standards, incident reporting and transparency, clear service-level commitments, and resilience and redundancy requirements. This is most plausible where the service is a true dependency layer. Cloud, payments, and certain logistics layers are common candidates.
4. Resilience policy for critical supply chains
Resilience is not free. If society wants redundancy, someone has to pay for it. Policy tools include strategic reserves where appropriate, multi-sourcing requirements for critical goods, incentives for geographically distributed capacity, and stress testing and scenario planning for dominant nodes.
This agenda should be narrow and evidence-driven. It should target high-consequence chokepoints.
5. Governance reform to reduce short-termism
Some concentration dynamics are fueled by financial incentives. Governance reforms can include long-term incentive alignment, stronger disclosure of operational dependencies, and limits or conditions on buybacks during periods of underinvestment in resilience. These are contentious. They are still relevant if the core concern is systemic fragility.
6. Public capacity for enforcement
Antitrust and resilience policy require capacity. That means stronger budgets for enforcement agencies, more technical expertise, faster adjudication pathways, and better coordination across regulators. Without capacity, the system defaults to reactive crisis management.
X. What will not work
Several common responses are insufficient. Voluntary restraint is unlikely to scale because incentives favor growth and consolidation. Consumer choice alone is insufficient because choice is limited when bottlenecks are concentrated. Small tweaks often fail because market structure frequently requires structural remedies. State-by-state approaches tend to fail for national systems because many markets are national or global.
The reality is that concentration has become a governance issue, not only an economic one.
XI. Conclusion: markets need competition, and systems need redundancy
Scale can create value. It can also create fragility. When corporations become infrastructure, the economy inherits a new category of risk.
Failures ripple. Costs are socialized. Accountability becomes unclear.
A modern policy agenda should treat concentration as both a competition problem and a resilience problem. That means enforcing competition rules where markets have become non-competitive, reducing conflicts of interest where platforms set the rules, requiring redundancy where single points of failure have high public costs, and building public capacity to oversee systems that the public cannot easily replace.
This is not anti-business. It is pro-competition and pro-resilience. It treats economic infrastructure with the seriousness it requires.
Sources
U.S. Department of Justice, Antitrust Division, Herfindahl-Hirschman Index (HHI thresholds; updated Jan 17, 2024): https://www.justice.gov/atr/herfindahl-hirschman-index
Federal Trade Commission and U.S. Department of Justice, Merger Guidelines (2023) (final PDF): https://www.ftc.gov/system/files/ftc_gov/pdf/2023_merger_guidelines_final_12.18.2023.pdf
OECD Competition Committee, Roundtable on Conglomerate Effects of Mergers: Background Note (DAF/COMP(2020)2): https://one.oecd.org/document/DAF/COMP(2020)2/en/pdf
De Loecker, Eeckhout, and Unger, The Rise of Market Power and the Macroeconomic Implications (NBER working paper version): https://www.nber.org/system/files/working_papers/w23687/w23687.pdf
Autor, Dorn, Katz, Patterson, and Van Reenen, Concentrating on the Fall of the Labor Share (AER Papers and Proceedings, 2017; working paper PDF): https://www.nber.org/system/files/working_papers/w23108/w23108.pdf
OpenSecrets, Federal lobbying totals (use if you keep quantified lobbying claims): https://www.opensecrets.org/federal-lobbying
Additional reading
The Rational Moderate
External
DOJ/FTC Merger Guidelines: https://www.justice.gov/atr/merger-guidelines
OECD competition note on conglomerate effects: https://one.oecd.org/document/DAF/COMP(2020)2/en/pdf
OpenSecrets federal lobbying totals: https://www.opensecrets.org/federal-lobbying
Leave a Reply