Published on January 29th, 2017 | by Donald McIntyre


Tokens, Circulation and Trust Minimization: An Explanation of the Kula Ring by Nick Szabo

Editors note: Relevant to blockchain tokens, ICOs and ensuing circular economies, the following 2005 essay by Nick Szabo explains the Kula Ring, an economic ecosystem in Melanesia that used collectibles as tokens or money to conduct trade:

An Explanation of the Kula Ring

Copyright (c) 2005 by Nick Szabo

(Some of the following previously appeared as a sidebar in Shelling Out — The Origins of Money and on the author’s blog, Unenumerated).

Abstract: A variety of vague cultural explanations have been proposed for the unique Melanesian trading instituion, the kula ring. Landa[L94] proposed an explanation based on credit risk and gossip. This working paper provides a rigorous explanation based on the author’s theory of collectibles and specific cycles of circulation as detailed in Shelling Out — The Origins of Money and using minimal trust assumptions.

Just off the cost of New Guinea, Melanesians evolved (over thousands, and perhaps even tens of thousands, of years) a unique commercial institution known as the “kula ring” for the collectibles that circulated within it. The unforgeably scarce kula collectibles doubled as “high power” money and mnemonic for stories and gossip. Many of the goods traded, mostly agricultural products, were available in different seasons, and so could not be traded in kind. Kula collectibles solved this double-coincidence problem as an unforgeabaly costly, wearable (for security), and circulated (literally!) money. Necklaces circulated clockwise, and armshells counter-clockwise, in a very regular pattern. By solving the double-coincidence problem an armshell or necklace would prove more valuable than its cost after only a few trades, but could circulate for decades. Gossip and stories about prior owners of the collectibles may have provided information about upstream credit and liquidity [L94]. In other Neolithic cultures collectibles, usually shells, circulated in a less regular pattern but had similar purposes and attributes.

Why do the kula collectibles circulate in a specific cycle? Why does the cycle traverse the ring of islands? And why are there not just one but two counterrotating cycles? The author for the first time (as far as the author is aware) provides a rigorous answer to these puzzles.

The proximate reasons why two kinds of collectibles circulated in opposite directions were transaction norms and religious taboos shared by the traders. The ultimate reasons (i.e. functional reasons why the institution evolved this way) may be explained as follows:

Collectibles often literally circulated in specific cycles that avoid sinks. This amortized the cost of the collectible across more trades. Why? The function of collectibles is to lower transaction costs such as non-coincidence of needs or events. A transaction network uses collectibles more efficiently if it increases the ratio of velocity to current value. Since velocity was very low and the value of transactions (trades, marriages, inheritances, etc.) could be high, this was a big problem in neolithic transaction networks. Some collectibles had a velocity of once per generation, i.e. they were just family heirlooms!

The network uses collectibles more efficiently with more members, so we’d expect the circulation to expand beyond bilateral cycles into cycles around the entire regular transaction chain.

Now that we’ve explained why a collectible cycle would be expected to arise in the first place, on to the key puzzle: why to counter-rotating collectible cycles rather than a single cycle around the island chain?

We can model the kula ring abstractly as a cyclical island chain of N islands, such that each island trades only with its immediate neighbors. These islands vary in their volume of trade. This can be model in the limit as N-1 islands of normal trade and an Nth island that trades nothing (other than, possibly, collectibles).

For illustration, take N=5. We thus have island chain ABCDE, where boats can travel no farther than their neighbors and island C is self-sufficient (or just very small) and doesn’t want to trade goods regularly, but AB, DE, and EA trade regularly. C breaks the collectible cycle (greatly decreasing the efficiency of collectibles for the other islands) unless it acts as a conduit. Simply taking collectibles from D and shipping them to B (and pocketing some profit of course) to complete a one-collectible cycle requires acts of faith. With two opposite circulating collectibles, C simply engages in bilateral trades with both B and D to complete both cycles. Thus, two counter-rotating cycles allow for full circulation with minimal trust assumptions.


[1] Landa, Janet, Trust, Ethnicity, and Identity: Beyond the New Institutional Economics of Ethnic Trading Networks, Contract Law, and Gift-Exchange, The University of Michigan Press, second edition, 1998.

Comments and criticisms to:

Tags: , , , , , , , , ,

About the Author

Founder of Naation. Inventor of Etherplan. Co-founder of Global Financial Access. Founder of Dineronet and McIntyre S.A. Previously, at Morgan Stanley and UBS.

Back to Top ↑