Network goods structurally resist competitive provision

For railways, telecoms, water networks, and electricity grids, the value comes from system integration and coverage, not from competition at individual nodes. The physics and economics of these goods push toward monopoly regardless of policy intent. The relevant question is not “is this market competitive?” but “is this monopoly well-governed?”

Explanandum

Why do attempts to introduce competition into network infrastructure (UK rail privatisation, AT&T breakup, water privatisation) consistently produce dysfunction, and why do these markets tend to reconsolidate?

Substance

Network goods have a defining characteristic: the value of participation increases with the number of other participants. A telephone network connecting 10% of the population is worth far less per connection than one connecting 90%. This creates powerful coordination pressures toward a single integrated system.

The historical evidence is remarkably consistent. Victorian railway competition produced wasteful duplication — rival companies building parallel routes, incompatible gauges, redundant stations. The market resolved this through consolidation and cartel behaviour, not sustained competition. Early American telegraphy went through the same cycle: dozens of incompatible systems, manual transcription at boundaries, eventual consolidation into Western Union. The telephone repeated it: after Bell’s patents expired, 3,000+ independent companies created incompatible exchanges, and the market reconsolidated.

Breaking a national network into regional pieces (Baby Bells, UK train operating companies) creates the worst of both worlds: no economies of scale AND no competitive pressure, since regional operators serve captive geographic monopolies.

Supports

• UK rail privatisation produced coordination failures, misaligned incentives (contributing to the Hatfield crash), and eventual reconsolidation toward Great British Railways
• The Baby Bells reconsolidated into AT&T and Verizon within two decades
• Early American telephone competition required businesses to maintain multiple subscriptions to reach all contacts — the market was value-destroying

Challenges

• Mobile telephony eventually introduced genuine competition to landline telecoms — but through substitution from outside the network, not competition within it
• Some network goods (airlines, internet service) do sustain meaningful competition despite network characteristics
• The “natural monopoly” framing can be used to justify regulatory capture and complacency

Open Questions

• Is there a principled way to distinguish goods where competition is structurally viable from those where it isn’t, before the experiment is run?
• Can designed institutional structures (regulated utilities, public ownership with independent oversight) consistently outperform the market’s own tendency toward consolidation?
• Do digital platform monopolies (Google, Meta) exhibit the same network dynamics as physical infrastructure, or are there meaningful differences?

Source Context

Developed through discussion of the AT&T breakup, UK rail privatisation history, and the observation that Albrecht’s framework has no way to address markets where competition is structurally unviable.