You are here

Mutual credit exchange and commodity money exchanges.

The Credit Commons is my design for a financial exchange based on the principles of mutual credit, and autonomous communities voluntarily and trustfully collaborating. As such it differs from the more common exchange architectures in fundamental ways. This post explains how and why there is a difference, and hence why mutual credit exchange is not well suited to current exchange architectures, and hence why the mutual credit exchanges in general, and the credit commons in particular, shouldn't just be grafted onto some existing technology.

A normal exchange:

  1. assumes that all assets are priced purely on supply and demand. They require the highest trade volume possible (liquidity) to reduce volatility and slippage.
  2. is a simultaneous swop of assets at the specific moment when the minimum price of the seller overlaps with the maximum price of the buyer.
  3. is a trusted institution. Traders load their assets into the exchange (which is why I'm calling it a commodity money exchange); the exchange re-allocate the assets to different account as trading takes place; and then traders pull their assets out.

Another kind of exchange, which I don't want to go into here because it only involves cryptocurrencies and seeks to further reduce the need for trust in any institution, is the decentralised exchange (DEX). in this configuration, the exchange doesn't hold assets on behalf of traders, but matches bids and offers, and executes atomic (meaning simultaneous and entangled) transactions on both of the blockchains involved.

In contrast, no assets go through a mutual credit exchange and the exchange that happens is not simultaneous. Participants in a mutual credit exchange trust each other to complete the exchange later. The ability of participants to trust each other is a major advantage because it means it enables half the users to buy before they sell, without having to acquire assets first, and it enables prices to be more stable, because there is no intermediate commodity (money) whose value can also change.

This difference between types of exchange is exactly the same as the difference in different types of money. Is money an asset which is used as an intermediate asset to 'make' markets, or is it merely a standard of value to help price things? Is an exchange the swap of money for a commodity, or is the swap of a commodity for a commodity?

I'm not talking about anything esoteric. I'm sure that there must be a vocabulary in economics for these types of exchange, with and without a commodity medium of exchange. This latter type of exchange is what Keynes proposed for international trade at the Bretton Woods conference, and what many would-be reformers of the global financial architecture are still calling for. But with the growing dogma trade is a process that happens between anonymous parties purely on the basis of competition over price, commodity money and commercial commodity exchanges are becoming more prevalent, leaving no headspace for the kind of exchanges involving trust, community, self governance, and p2p credit.

It is possible to build a mutual credit exchange on a normal exchange architecture, but it is hacky. First of all you have to create assets (tokens) out of nothing in the central exchange, give them a face value and give some to the participating members. This represents the credit limit of each member, the maximum they can spend over what they earning. Then you need to write a script or somehow prevent members from having too high balances - a different mechanism since you can't have negative assets. Traders are not interested in a 'marketplace' where each commodity is bought and sold for 'money' because there is no 'money' and there is no instantaneous swap of assets. Instead traders want to effectively transfer their worthless tokens to another user (or claim them from another user). Every transaction should be reversible in case or errors or disputes.

With this hacky approach I'm assured that we could use the Bancor protocol to build a mutual credit exchange. The Bancor Protocol is a simple smart contract that manages a pool of assets so as to guarantee a counterparty in markets with low liquidity. a mutual credit exchange is considerably simpler, and does not even exhibit the problems that the Bancor Protocol addresses.

We might get further by using Ripple/Stellar, although they model networks of relationships not groups and groups of groups. Consequently its very hard to model groups as objects with properties (like transaction validation rules) in that architecture.

But there's another reason we need a dedicated system. Because mutual credit systems are for people who trust each other, you can't meet the needs of everybody by just building one and putting it on a really big server. Instead you need to build many and link them together. Ideally each group (and group of groups) would have its own software instance and be independently hosted. In order to do one transaction across the network, many software instances could be involved, and they would need to coordinate. If the community currencies doing mutual credit were to to use exchange software build in the commodity money paradigm, this kind of separation and connectedness of groups would be extremely elaborate and hence expensive to build and maintain.

What we need is something simpler that a standard exchange. If what we need is two bicycles we won't thank anybody for giving us a car!


Isn't the easy and simple answer to this problem to use holochain, which is by its nature person oriented rather than coin oriented?

I've thought about this a lot and the problem of exchange is that it is fundamentally about interoperability between systems. Holochain is a system. So the credit commons needs to be a protocol for systems to talk to each other, not a system in itself. I don't want to say to everybody, if you want interoperability you must use this software or this blockchain or whatever, but you must extend your software to implement this protocol.

Add new comment

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer