Stateless Somalia

When Somalia’s central government collapsed in 1991 it did not become a controlled experiment in liberty, but it did become the closest thing the modern record offers to a comparison of a population under a predatory state versus the same population without one. Peter Leeson’s verdict from that comparison is blunt: the government “did more harm to its citizens than good,” and its disappearance opened room for progress. This is the empirical counterweight to the intuition that any state, however bad, beats anarchy — and it is the case the wiki’s war-and-state arguments point at when they call the state predatory rather than protective.

The finding

Leeson assembles eighteen welfare indicators that allow pre- and post-collapse comparison and reports that on nearly all of them Somalis improved:

on nearly all of 18 key indicators that allow pre- and post-stateless welfare comparisons, Somalis are better off under anarchy than they were under government.

The mechanism he credits is not anarchy as magic but the removal of predation: with the extractive apparatus gone, he attributes the gains to renewed economic vibrancy and “public goods in the absence of a predatory state.” Telecoms, livestock export, and money-transfer networks grew under customary law (xeer) and clan-based dispute resolution rather than statute.

The conditional, stated plainly

The claim is conditional, not utopian. Leeson’s own framing is that the result depends on how predatory the prior state was: where a state is highly predatory and unchecked, government can fail to add to welfare and instead push it below its stateless level. Somalia qualified; not every state does. The argument is therefore narrower and sharper than “anarchy is always better” — it is a sufficiently predatory state is worse than its own absence, which is exactly the regime Political Means and Economic Means and State Power and Intervention describe.

Scope

This is descriptive development economics, not anarcho-capitalist political theory: absolute development stayed low, the order was customary rather than designed, and the post-2006 record (foreign intervention, the rise of armed Islamist movements) is a separate chapter. Its value here is as evidence, not ethics — it is the data point that keeps the state-as-predator thesis from being a pure deduction. It also marks the boundary where War and State Formation’s descriptive account of how war builds states meets a case running the other direction: state collapse, then order without the state.

See Also

Sources