Note:32 On The Agency Problem at the Core of Internet Governance

What appears today as ideological hostility toward commerce in internet governance is not ideological at all. It is structural. The dysfunction originates from a textbook agency problem: the global internet was placed under the control of technical agents who do not own, do not invest in, and do not economically benefit from the system they govern. Predictably, they optimized for power retention, role preservation, and institutional relevance rather than value creation.

Internet governance institutions were never designed around ownership, risk, or capital formation. They were designed around coordination. Once coordination bodies acquired allocation authority, scarcity control, and enforcement leverage, coordination quietly transformed into command. Not because of malice, but because agents maximize what they can measure and retain: influence, jurisdiction, and indispensability. Markets reward growth and efficiency; bureaucracies reward control and continuity.

Technical professionals, by nature of their career incentives, prefer stable rule systems that elevate procedural legitimacy over outcome efficiency. They rotate between registries, policy bodies, consultancies, and compliance roles. Their human capital appreciates when institutions become more complex, more opaque, and more centralized. Simpler systems reduce their bargaining power; competitive markets reduce their gatekeeping role. This creates a natural bias toward administrative allocation, artificial scarcity, and moralized justifications for intervention.

The absence of ownership is decisive. These agents do not bear downside risk when innovation is slowed, when capital formation is suppressed, or when infrastructure markets stagnate. They do not participate in upside when efficiency improves. Without equity, profit share, or balance sheet exposure, there is no internal signal telling them that growth matters. In such a system, economic sufficiency is not a goal; procedural dominance is.

This is why internet resource allocation increasingly resembles centrally administered distribution rather than market exchange. Scarcity is treated as a moral resource to be rationed, not an economic signal to be priced. Allocation becomes conditional, revocable, and politicized. The language shifts from utility to stewardship, from productivity to “community interest,” from efficiency to “fairness,” defined internally by the same agents who benefit from discretion.

Importantly, this outcome does not require ideological commitment to anti-capitalism. It emerges automatically whenever governance is detached from ownership. Engineers are trained to minimize failure modes, not to maximize returns. When placed in control of economic infrastructure without corrective incentives, they predictably over-constrain, over-regulate, and over-centralize. What looks like hostility to business is simply indifference to business outcomes.

The internet, however, is no longer a research network. It is a capital-intensive, profit-driven, global production system. Treating its core resources as administratively rationed inputs is economically incoherent. Capital markets cannot function efficiently under revocable rights, ambiguous tenure, and discretionary enforcement. Innovation slows not because of technical limits, but because governance suppresses predictable return.

Control must therefore migrate back to those with direct economic exposure. Shareholders, CEOs, founders, and operators whose wealth, valuation, and survival depend on business performance are the only actors structurally aligned with long-term value creation. Unlike agents, they cannot externalize failure. Unlike committees, they cannot indefinitely defer consequences. Their incentives are legible, measurable, and enforced by markets.

This realignment is especially urgent given the nature of IPv4. IPv4 addresses represent a once-in-a-generation opportunity: the first globally recognized, mathematically finite, non-replicable digital scarcity to underpin an entire industry. Unlike spectrum or land, IPv4 scarcity was not politically constructed; it is protocol-native. It is already accounting for double-digit percentages of market capitalization in some telecommunications firms, and its economic gravity is still increasing.

Allowing such a foundational scarcity asset to remain governed by agent-dominated, non-owning bureaucracies is not merely inefficient — it is value-destructive. The future of internet infrastructure depends on recognizing IPv4 not as a moral resource to be administered, but as capital to be allocated by those who bear the consequences of allocation decisions.

Until shareholders and executives with real balance-sheet exposure take control, internet governance will continue to drift toward centralized administrative logic. Not because of ideology, but because agency problems, left unresolved, always converge on power over value.