Tokenizing the mineral supply chain

Our previous eight blog posts have shown the mineral supply chain is full of complex problems that have multi-faceted causes.

Blockchain, paired with careful system design, provides us with tools to tackle these problems.

Blockchain has four key characteristics that make it suitable to address these increasing regulatory challenges. In the upcoming posts, we will explore these characteristics in detail and build up a model for tracking the mineral supply chain.

The Technology

The first two characteristics we will look at are the core results of using blockchain. They are tools that allow us to create a system that can keep track of the history or provenance of the materials we are transacting and ensure that record is trustworthy:

1. Blockchain can tokenize minerals and metals, along with key due diligence data submitted by all participating supply chain actors. These tokens are digital assets that are highly tamper-resistant.
2. Blockchain maintains an unchangeable record of transactions for these tokens. This gives us the ability to create an immutable chain of custody and confidence in the data in our system.

The Incentives

The next two characteristics provide incentives for every supply chain actor to use our well-designed technology tools. They are how we can turn our system from a novelty for companies who know each other well, into a global infrastructure that can address some of the deepest problems the mineral supply chain faces:

1. Blockchain can create self-governed incentive systems, to make it beneficial for supply chain actors to participate in the system.
2. Blockchain allows parties who do not trust each other to transact without a trusted third party, eliminating the need to trust a single company with supply chain data.

The success of our mineral blockchain tracking system depends completely on designing the technology tools properly, as well as the incentives.

Let’s start by delving into the tools of the trade in this blog.

The Technology: Using tokens to create an unchangeable record of supply chain history

When most people think of blockchain they think of tokens. Tokens are digital assets that are created according to rules agreed upon by everyone using the blockchain. They cannot be produced in any other way, and cannot be duplicated or forged. This property means we can limit the number of tokens available, and this rarity means they can represent real-world assets such as minerals or metals.

If we are going to use a token to track our minerals, we first need to decide if we want to use a fungible or a non-fungible token for the task. Fungible tokens are all alike and interchangeable, similar to cash in a bank account. Non-fungible tokens are all unique, like collectable trading cards. The decision comes with tradeoffs no matter which we choose.

Fungible tokens

The simplest type of token we can create is a fungible token. With fungible tokens we only keep track of the quantity we have; every token we purchase can be merged into a larger pile of tokens making them all indistinguishable from each other.

Fungible tokens are the most commonly created type of token in the blockchain world. They are easy to handle, like a currency, can be stored in a wallet as a number, and moved around like cash. Metals are also fungible so it seems intuitive that a fungible token would work for tracking responsible materials.

So, if we tokenize a supply of known fungible metal using a fungible token, how could it work?

Example 1 — Fungible token representing gold

Imagine we start a company that wants to produce responsible gold for investment purposes. They purchase all their gold from one large scale refiner that purchases all their supply from one large scale mine. All of the production practices are responsible and are regularly audited so our company can be confident that the gold is responsible.

We purchase this gold in the form of 1oz bars and put the bars into a vault. For each bar in the vault we create a fungible, digital token on a blockchain which is then sold on the open market. This gold token would contain a link back to a website that shows how responsible the gold is. At any point, the owner of a gold token could redeem the token with our company to receive a gold bar from our vault.

Positive Outcomes of this Approach

• Fungible tokens are easy to manage because we do not need a physical identifier linking the token to the metal. We know all the metal is in a vault, and it all comes from one source, so it doesn’t matter which bar is which.
• Investors can move into and out of gold positions fluidly with minimal transaction fees.
• Implementation is straightforward — there are few real-world company processes that would have to be changed, and only one company (ours) needs to know how to use the blockchain. Our company can decide to implement such a system unilaterally without the buy-in from every member of the supply chain.

Limitations of Fungible Tokens

• The tokens are only able to be fungible because we control all the aspects of responsibility of the supply chain and keep the gold from these processes separate from the rest of the market. It works well because we have one vault storing all our gold coming from only known sources and audited supply chains. If multiple companies each create their own vault and their own fungible token, these tokens could not be mixed without losing the information about which vaults held our gold. This makes such a system difficult to scale
• Gold token buyers have to trust us when we say that the gold is responsible. Our company is providing data on a webpage showing that the gold in our vault is responsibly sourced, but there’s nothing in the system itself that tracks the gold from the extraction point up the supply chain. This might be good enough when we have one source and a lot of data on that source, however to scale, we will likely need stronger supply chain traceability.
• Fungible tokens may represent responsibly sourced metals, but they are still very valuable digital assets that can be transferred at will between wallets without needing a middleman. Ironically, this means our responsible gold tokens could be used to launder money, and we would need to take measures to prevent this.

Example 2 — Adding a layer of complexity to fungible tokens representing gold

Our previous example had very limited use of blockchain. We were able to create a digital representation of the gold in a vault, but the gold didn’t move. We can think of this as being able to track the amount of gold in the vault, which is one actor at one level of the gold supply chain.

Can we just add actors each with their own fungible tokens to get better depth to our supply chain tracking?

Imagine a gold mine processing ore onsite into “doré” bars, which is partially refined gold that still has silver mixed in. For each bar, they create a doré token, which is fungible. Just like the example above, the tokens are not traceable to a particular bar, but rather the amount of tokens are limited by the quantity of doré produced on site.

When a shipment of bars is sold to a refiner, the mine would send an equivalent amount of doré tokens to the refiner. The refiner would then be able to destroy the doré tokens in order to create gold bar tokens.

The gold bar tokens would be sent to our company along with the gold bars which are then put in the vault. With the gold bars in the vault, our fungible gold bar tokens can then be sold on the open market as in the example above.

This system will allow us to track volumes of doré produced at the mine and volumes of doré that are converted to gold bars, however it does not allow us to track specific bars or shipments.

Positive Outcomes of this Approach

• This method gives us data about how much material is coming from each source, and this data is stored on the blockchain.
• The tokens at the end of the process are still fungible and that makes them easy to trade.

Limitations of this Approach

• Because we are tracking materials along a supply chain, we need all companies along the supply chain to participate, making it more challenging to implement a project. These companies will likely need to adjust their internal processes in order to keep track of these blockchain-tracked shipments, meaning more work and cost.
• Each new supply chain actor who creates a fungible token is an actor we need to trust. This means there are multiple points where the system could be vulnerable to human error or malicious intent because we now have to record more than just transactions, we have to record transformations (when the metal is refined and processed).
• Because the tokens are still fungible, we cannot link them to specific shipments of metal. This limits our ability to record the transformations because we don’t have data on the specific shipments.

This method essentially introduces a level of non-fungibility into our system. Although both the doré and the gold bar tokens are fungible, they cannot be mixed with each other.

By adding layers of non-fungibility we get greater visibility into what’s going on in our supply chain, at the cost of it being more complex to implement.

Non-Fungible tokens

Now that we have designed a basic token-based tracking system, we want to have an even greater level of detailed information which metals come from where. Perhaps traceability is more important to us than trading liquidity. For this, we can use non-fungible tokens to represent metal in our supply chain.

Non-fungible tokens can be created to represent specific shipments of metal ores or ingots, so they are not interchangeable with another token, and effectively act as a digital twin that can be used to demonstrate who owns it. Each token can contain a link to data that demonstrates the responsibility of the shipment that it is linked to.

Example 3 — Non-Fungible Token Representing Gold

We can use a non-fungible token in our gold scenario which will act a bit like passing a baton from person to person in a relay race. When it gets to the end, it will have the fingerprints of everyone who held it, giving the owner a detailed history of the item.

To illustrate, let’s go back to our gold mine again. Our responsible gold mine wants to sell responsibly sourced gold to their customers.

They extract gold, melt it into doré bars that are weighed, and a non-fungible “doré” token is created to represent each bar. This time, the doré token contains a unique ID, which is put on a tag and attached to the doré bar, as well as a link to a database of due diligence data about the company that produced it. As the doré bars are passed from owner to owner, the associated non-fungible tokens are also passed along to each new owner.

When each doré bar reaches the refiner, like before, it is processed into 1 oz gold bars. Unlike before we will not destroy the doré tokens, but rather freeze them so they can no longer be traded. When the new gold tokens are created they will link to the old doré tokens so we have a record of where they came from. When we put our gold bars in a vault, each will have a unique ID and customers can sell them on the open market, or redeem them for their specific gold bar.

Improvements Over Fungible Tokens

• Non-fungible tokens allow us to represent individual units of material (the doré bars and the kilo bars). Having this extra granularity of data means we are not limited to a single mineral source or vault.
• Because the individual shipments are represented, we can also record who held them and where they were produced. This allows us to track minerals from the point of production, all the way down the supply chain.
• The tokens are linked to specific bars or shipments, making them more difficult to use for money laundering.

Some New Challenges

• For the tokens to track the metal, we need to link the metal to the non-fungible tokens. Tagging presents a challenge whenever the material is processed or changes form.
• We still need to ensure that good data is provided to the blockchain. Our non-fungible tokens are only as useful as the data they are based on.

And of course, we lose all the benefits of fungible tokens. If we are tracking individual gold bars with a unique ID, we need to check these IDs whenever we trade them and make sure we have the right ones if we sell the physical bars. There is no getting around this. The point of tracking individual bars is to de-commoditize the gold.

If we want to say that a gold bar that comes from a responsible source is different than one that funded human rights abuse, this is the trade-off.

It is important to note that the methodology that we have built up is not the only way to apply blockchain to address issues in the mineral supply chain. If you’ve read the Minespider white paper, you probably noticed that it describes a system that is a bit different than what we described here, and that is because it was designed for a more complex and more generalized scenario.

The main goal of this blog post is not to completely solve the problems of the mineral supply chain but rather to get familiar with the tools that blockchain gives us, and to begin to think through the implications of different ways of applying them in a systematic way.

In this post, we looked at how to build up the technology in scenarios where we had a lot of implicit trust between the supply chain actors. In our next post, we will look at different ways we can use blockchain technology to address the points of mineral and metal processing in larger scenarios where we might not have as much trust between players.