04 Jan 2018

A token model

In 2017 the Blockchain world bloomed with many token offerings. The general idea behind these is that innovative networks could get some kind of seed money by selling initial tokens to investors. These Utility Tokens are designed to be currencies in these networks and their creators argue that they should be valued using the equation of exchange.

As much as I like the idea of mathematics to guesstimate value, I also love to “grasp” the actual behaviours which trigger transactions.

On this front, one of the most convincing arguments that I found for why Bitcoin is valuable (not necessarily to its current levels though!) is that it is the most convenient medium of payment for a lot of illegal activities. Bitcoin is the facto currency for anyone who buys illegal substances online.

As long as people will be addicted and that no other form of payment will provide orders of magnitude more convenience (safety, relative pseudonymity, speed…), people will use and need Bitcoins. Similarly, even in countries with mind boggling inflation, as long as people have to pay their taxes in the local currency, most people will want to be paid with it too.

Using a similar reasoning, token creators often claim that their currency is valuable because it is required to transact on their networks. As such, and assuming that the network provides a valuable service, people will want to own (and hence purchase) these utility tokens.

Unfortunately I have serious doubts, like others (read Kyle Samani or John Pfeffer thoughts on this), that each of these tokens, or even any of them can actually be treated as a currency. In the same way that casino tokens are required to play slot machines in Las Vegas, the room and brunch buffet can only be paid for in dollars.

How can we then expect the crypto tokens value to appreciate in order to bring the ROI expected by ICO investors?

In a recent exploratory work that I did for Medium, I designed a token which would be attached to a dividend, paid in Ether. The dividend would be paid from a tax captured on each token transaction.

Let’s assume that a network exists and provides a valuable service. The service is executed by a smart contract on the Ethereum public blockchain. Performing or receiving the service is not tied to ownership of the token. However, since the service consists of Ethereum transactions, fees have to be paid to the Ethereum blockchain. We could easily add that an additional fee has to be paid to the contract itself (on top of the Ethereum miners). This fee is then collected in a shared bucket in the contract and each token holder can, at any point, withdraw their share of the dividend from that bucket.

The benefit of this approach is that (at least for a fixed supply of tokens), the token’s value will increase with the actual network usage. We can use basic DCF to calculate how much each token is worth.

The most immediate criticism is that adding a fee to the network will slow down its growth. However, I think this is missing the point that it would then be in the best interest of any participant to own their relative share of activity on the network so that the amount of fees they pay is offset by the dividend they receive.

If a single participant is responsible for 10% of the network activity, they will indeed pay 10% of the fees. If they also owned 10% of the outstanding tokens, they would also collect 10% of the dividend, balancing their fee contribution.

This model is in fact very similar to traditional equity models. The Smart Contract can be seen as a corporation performing a service and the fees are the costs of capital, paid as dividend. In this model, tokens are shares which can be sold to investors who have a strong belief that the network will perform a valuable service. The one difference resides in the fact that in this crypto world, customers are strongly encouraged to take ownership of the services they consume. By doing so, they can reduce their acquisition price and capture some of the value they create by using the service.