Thread by @fintechjunkie

As promised - more red pill insights (Week 4)!

My crypto journey continues with an unpacking of what it would take to replace a network-based transaction (Visa/MC) with DeFi rails.

Spoiler: Moving money is the easy part. Detailed🧵👇

A foundational question worth asking is:

“Why replace the networks?”

Billions of cards are in force that are accepted by tens of millions of merchants globally. It isn’t free and merchants don’t get paid instantaneously, but the system works.

In North America alone, about $0.5T of revenue is generated by payments with a third attributable to credit card rails. The numbers vary in other regions of the world, but the commonality is that merchants consider interchange a major cost that they’d like to see reduced.

If the only functionality the networks’ rails provided was to move money from one ledger (a consumer’s bank account) to another ledger (a business’s bank account) then the interchange fees could justifiably be classified as “highway robbery”.

DeFi rails can move “good funds” in a trustless manner without the networks and without any participation from a Banking institution.

But this completely misses the point that the established networks do more than just move digits from one ledger to another.

A critical value the networks provide is the creation of trust.

They explicitly manufacture trust in the system such that consumers are willing to pay for goods/services in advance of delivery and in advance of confirmation that the goods/services are acceptable.

The networks set the rules that merchants and issuing banks need to follow. The networks also play a policing function to enforce the rules.

Rules + Enforcement = Trust.

The net result is “a confident relationship with the unknown” which greases the wheels of commerce.

So when consumers see that a merchant accepts Visa/MC/Discover/Amex, they have some level of confidence that the merchant won’t cheat them. And if they run into a problem with a merchant, there’s a belief that their Card Issuer will step in to resolve the issue on their behalf.

This is only possible because each merchant that wants access to the network rails needs to pass a rigorous underwriting process which is designed to ensure that they are honest, financially viable, and meet certain basic “customer rights” standards.

The underwriting process isn’t a simple “fog a mirror” tick-the-box exercise. And the rules are designed based on the first principle that a consumer’s complaint is accurate and justified until proven wrong by the merchant.

The rules are complex. For instance, here's Mastercard's 798 page chargeback guide:
mastercard.us/content/dam/mc…

But they result in a “zero fraud liability” guarantee and the ability for a consumer to dispute any transaction at any time.

So the operative question is: Can DeFi primitives be assembled that replace the networks’ “trust based system” with a less expensive/more efficient “trustless system”?

Let’s imagine what a “trustless shopping experience” could look like:

For a trustless experience to work, it has to overcome at least five major pain points that are addressed by trust in today’s system.

Ownership: Trust that the merchant has access to and ownership of the goods they’re selling

Fair Exchange: Trust that the merchant will ship the goods once payment has been authorized

Accurate Representation: Trust that the item will be received in the represented condition

Returnability: Trust that the merchant will adhere to its stated returns policy

Disputes: Trust that disputes will be resolved quickly with the burden of proof on the merchant

It’s difficult to imagine satisfying all five points of trust in a completely trustless manner, but it is easy to imagine a system that moves trust from the individual merchant to an entity that is designed to act as an agnostic intermediary built on top of smart contracts.

Imagine a system that starts with merchants moving their goods to an intermediary who’s responsible for storage, shipping and returns. This intermediary can ensure that each item they’re in possession of is “tagged” such that smart contracts can govern their actions.

Each item could be assigned three NFTs:

Item NFT: Data and pictures that describe the specific item and its provenance

Ownership NFT: Held by the owner of the item

Location NFT: Data that describes an item’s current location as well as its intended destination location

Primitives can be built on top of this NFT structure that handle very specific tasks without requiring trust.

For instance, when an item is purchased, a smart contract can be created that does the following:

💥Moves tokens from the purchaser’s wallet to the intermediary’s wallet

💥Modifies the destination location to the purchaser’s chosen destination

💥Executes a packing and shipping primitive

The item’s Location NFT can be modified in transit, and when the current location = destination location the item has been delivered and the "return policy" clock starts ticking.

The buyer can review the item and accept it which would execute the following primitives:

💥The Ownership NFT would be modified to make the purchaser the new owner

💥Tokens would be moved from the intermediary’s wallet to the seller’s wallet

Or the buyer can review the item and decide to return it which would execute the following primitives:

💥Modification of the Location NFT to set the intermediary as the new destination address

💥A shipping primitive for picking up and returning the item

And when the item is returned to the intermediary, the following primitives would be executed:

💥Intake, inspection and reshelving/disposal

💥Modification of the Item NFT to reflect its provenance (where it’s been) as well as its new condition (opened box) inclusive of new pictures of the actual item

💥Moves tokens from the intermediary’s wallet to the buyer’s wallet

Primitives could be created to address situations like “lost in transit” and “disputes”. Sellers could get paid instantly by purchasing “returns insurance” which could be backed by 3rd party stakers. Layers of buyer and seller value props can be built on top of this structure.

But is this worth the hassle? It depends on your goals.

Today’s system is designed to use underwriting and policies to REDUCE the instance of broken promises downstream and REMEDIATE bad outcomes. Transactions where promises are kept create very little/no work/cost.

A trustless system incurs incremental costs for ALL transactions in its attempt to ELIMINATE broken promises downstream.

It isn’t clear what the cost/benefit analysis would reveal if today’s trust based system were compared to a system designed to be trustless.

My intuition screams that it will take many iterations and additional technological advancements (IOT? More efficient Blockchains?) before a trustless system can compete economically with today’s trust based system.

But the experiments are worth running and will be fun to watch!