Witnet 2.0 discussion: tokenomics

Witnet 2.0 encapsulates a lot of changes to the protocol, and tokenomics is part of the change. This article will dive into some of the potential changes we could see working within Witnet.

Rokowski
The Witnet Oracle Blog

--

Introduction

There are a few disclaimers here that we want to make before getting into the actual figures for these four proposals. We announced these tokenomics changes in our May community call back on the 26th. You can read the announcement here.

First and foremost: we must remember that $WIT is a utility coin. It is not a store of value or medium of exchange like Bitcoin. That means the security budget is much different. Witnet has value because people are willing to pay for the security it offers; people pay something of value and are returned that value in the form of secure data feeds. Just as you would pay to have your transaction secured on Bitcoin or Ethereum, you pay for a data feed on Witnet. This is important to highlight because we are trying to make the protocol future proof rather than create a short term price increase. This is evident by the fact that we are having discussions with the community and not just simply changing the issuances to the lowest feasible inflation rate (it wouldn’t be possible anyway due to Witnet’s decentralized nature). In other words the most aggressive tokenomic changes aren’t necessarily the best for the protocol and these decisions are not based on valuation increase in the short term. If you are a community member looking to partake in the discussions please keep this in mind. Changes aren’t meant to turn Witnet into a deflationary network, as the block rewards still provide strong incentives being that Witnet is still very young, and the burning mechanism that would likely be implemented as part of Proof of Stake would not be significant enough to deflate the supply.

Finally, these are just some changes we have been discussing internally and we encourage you to do research on your own and consider submitting a proposal for a WIP. Witnet is public and anyone can submit a proposal to change the protocol, albeit it doesn’t mean it will be accepted by the network.

Issuance change one: do nothing

This is the simplest proposal for Witnet. Of course, it would change nothing about the protocol, but it would be the easiest to “implement” and wouldn’t need community support.

This would keep the total maximum supply at 2.5 billion with 250 $WIT per block that halves every 5 years.

As we already know, there is little support for this and we are looking to change it, but it still remains important to leave on the table as node runners could down-vote any change and keep the protocol the same.

Keeping the protocol the same is still a net-positive because the system works right now and adding changes can have unforeseen consequences.

Issuance change two: shorten halving to 3 years

This is by far the most aggressive change that could be implemented. Put simply, we could keep the same halving dates but instead of five years it would be three.

This would change the total maximum supply from 2.5 billion to 1.8 billion, a 28% reduction in total supply.

This is a simple change to the protocol that would probably have a lot of support. The drawbacks of this proposal is that the future miners (or validators) could be hurt, because if this passed there would only be another 700 million $WIT left to be mined.

Issuance change three: 8% inflation, decay by 15%

The third proposal that could be fitting for Witnet as a utility, would be to set yearly inflation to 8% upon approval of the WIP that would decay by 15% a year. This would essentially put the supply at 2.27 billion $WIT in the year 2050, but would take out any maximum supply. Beyond the year 2050 any issuance would become quite flat.

This would essentially take just under 225 million $WIT from the total maximum supply, which amounts to be a reduction of 10%.

This is an intriguing proposal because it’s a predictable monetary policy and the market can determine the fee rates around things like value transfers and data feeds, because those don’t affect maximum supply but will make for significant miner revenues in the future. Additionally, this proposal can be good for the Witnet security budget because there would essentially be tail emissions in a predictable way that allow for miners and validators to recognize how to organize their setups.

Issuance change four: 8% inflation decays by 25%

This proposal is similar to the third detailed above in that it begins with 8% inflation rate a year, and decays by 25% on a yearly basis. This would situate the total supply at around 1.93 billion around the year 2050, again a yearly issuance afterwards is pretty flat because of the decay system. By now you probably recognize that this is the middle ground between the most aggressive proposal (two) and the least aggressive (doing nothing).

Again, this proposal would make for a predictable monetary system that allows network participants to value their setups when contributing to network security. By 2050, fees make up the most significant portion of node revenue which is a sustainable system.

That’s it for now

These four proposals have been on the table for a while, but as we said in the early days of making Witnet 2.0 public, the tokenomics discussions took a back seat to Proof of Stake. At this point in time we have a good grasp on Proof of Stake and are ready to re-engage the discussions on the tokenomics portion of 2.0.

Connect with Witnet

Website | Twitter | Telegram | Discord | YouTube | Reddit

--

--