Scaling and Localism: Why Ethics Is Local, Not Universal

One of the persistent fantasies of moral philosophy is universalism: the idea that ethical principles that apply to individuals in small communities should apply equally to nations, and that what works morally between friends should work equally well between strangers across the globe.

Nassim Taleb argues this is empirically wrong, and explains why.

Ethics Doesn't Scale

Elinor Ostrom (Nobel Prize, Economics, 2009) demonstrated something counterintuitive: small communities reliably manage shared resources — fishing grounds, forests, water — sustainably and cooperatively, without government regulation and without privatization. Large groups managing the same shared resources typically fail.

The mechanism Ostrom identified: small communities where members know each other develop enforcement norms organically. Defectors — people who take more than their share — are identified quickly, ostracized, and denied future cooperation. The social cost of defection is real, immediate, and borne by people you interact with repeatedly.

As groups scale, this mechanism breaks down. You cannot know personally every member of a nation-state. You cannot enforce social norms against defection by people you'll never meet. The moral enforcement that works between neighbors through reputation, relationships, and repeated interaction doesn't have the same mechanism at national or global scale.

This isn't a failure of morality. It's a description of how morality actually works for humans: through specific relationships with specific people, not through abstract obligations to statistical collectives.

Taleb's Political Hierarchy

Taleb offers a clear hierarchy that maps his ethical commitments to scale:

"I am, at the Fed level, libertarian; at the state level, Republican; at the local level, Democrat; and at the family and friends level, a socialist."

This isn't incoherence. It's a recognition that different scales require different organizational principles because the ethical mechanisms that work at each scale are different.

At the family and friends level, he's a socialist because redistribution within small groups is natural, voluntary, and enforced by love, obligation, and relationship. Your sister is sick and you have money — you help. There's no question about it.

At the local level, democratic participation is meaningful because the effects of local policy are felt locally, by the people making the decisions. Accountability works.

At the state level, he moves toward more limited government because the accountability mechanism weakens at scale. Bureaucracies at state scale are harder to hold accountable than local institutions.

At the federal level, he's libertarian because the abstraction is complete: the decision-makers are fully insulated from the consequences of their decisions on the populations affected. The accountability mechanism is essentially broken.

The Abstract Other

The deeper point is about what "the Other" means at different scales.

When your neighbor is the Other — a specific, named person you interact with, whose wellbeing is visible to you, who can affect your reputation and social standing — your ethical behavior toward them is strong. You see their suffering. You know their situation. The cost of defection against them is real.

When the Other is a statistical abstraction — "the poor," "the developing world," "future generations" — the ethical pull is genuinely weaker. This isn't because people are hypocrites. It's because the ethical mechanisms that produce reliable behavior (reputation, relationship, repeated interaction, visible suffering) aren't engaged by statistical abstractions.

This is why charity to neighbors is more reliable and more efficient than international aid. The local giver knows the local recipient; they can monitor how the help is used; they have ongoing relationship; defection is costly. International aid operates through institutions that are multiply-insulated from the populations they serve.

The Decentralization Argument

This is why Taleb consistently favors decentralization over centralization. Decentralization maintains the ethical mechanisms that produce reliable behavior: skin in the game, visible consequences, real accountability.

The federal bureaucrat making decisions about a community she doesn't live in, whose residents she'll never meet, whose outcomes she won't experience — is operating in the most adversarial conditions for ethical behavior. She doesn't have the information, the feedback, or the accountability that would make her decisions reliably good.

The local official making decisions about a community she lives in, with neighbors who will talk to her at the grocery store about what's working and what isn't, who faces electoral accountability from people directly affected by her decisions — is operating in much better conditions for ethical behavior.

Decentralization keeps decision-makers inside the consequences of their decisions. Centralization separates them. This is the scaling-and-localism argument applied to governance.

The Implication for Global Ethics

This doesn't mean global ethical obligations don't exist. It means they should be appropriately weighted.

The strongest obligations are local: family, friends, neighbors, community. Weaker but real: regional and national. Weakest but still present: global.

The error is applying the moral intensity of local obligations to global contexts and designing institutions that attempt to enforce global ethical standards with the mechanism of local social norms. Those mechanisms don't transfer. What you get instead is institutional theater: organizations that perform global moral concern without producing it.

The more reliable path: strong local institutions that maintain real accountability, federated into regional and national structures that maintain accountability at their level, connected to genuinely minimal global frameworks where the stakes are too global for any local resolution.

For the full framework, read Skin in the Game Explained.