Death By Node Rewards?, Part 2

Why we should all calm down about node reward inflation

The idea of a node reward death spiral is all over the ICP cyber space recently because of the wild drop in ICP price to the $6 handle. I wrote on this topic back in December and while the analysis still holds up, I wrote it with the framework that single digit ICP prices were very unlikely. Yet, here we are. Let’s update the analysis to see if node rewards really are likely to cause an inflationary death spiral for ICP.

Background

A node reward death spiral goes like this: the Internet Computer runs on independent nodes, which have fixed costs. To incentivize these node providers to continue their service, the Internet Computer pays a monthly allotment in ICP to operators, who presumably need to sell the ICP to pay for their fixed costs in fiat. Since the node provider rewards are based on fiat terms, as the price of ICP goes down more ICP needs to be minted to pay the operators. If the operators then sell that ICP and the price goes down more, then more ICP is needed the following month. And so on until the price of ICP plummets to 0.

Data

Below is a chart of node rewards paid each month and the average closing price of ICP.

It’s clear that mid-2021 saw large outflows to node providers, most likely for reimbursement of their services prior to the token launch. Since October 2021, however, node rewards have ranged between $1.5M USD - $3M per month.

Analysis

Let’s take the high range of the above data and assume node reward will be ~$3.5M each month for the remainder of 2022 (the extra $0.5M to account for new node operators being added to the network). This is a worst-case scenario. If the price of ICP drops to $3.5, that would require 1M ICP to be minted to pay node providers each month. How impactful would that actually be? Compare that number to these facts:

• ~3.7M ICP are minted each month as governance rewards (albeit, some or a lot of that stays on the NNS due to merged maturity).

• As we enter Zone 2 there will be ~2.5M ICP less dissolving off the NNS as compared to December through April.

With this context, it’s clear that $3.5 ICP (1M ICP) would not be a death kneel for ICP, as the inflationary pressures would be on par with already existing inflationary pressures. In fact, I would argue that prices would have to full MUCH lower than $3.5 to kick off any inflationary spiral. In addition, this analysis neglects the fact that the Internet Computer could slow the addition of node providers, further decreasing inflationary pressure. It also neglects the fact that time is also needed for the death spiral to occur. If ICP dropped to $3.5 for a few months, but then rebounded, there’s hardly any impact long term (except the extra minting of maybe a few million ICP). It would take an extended amount of time at depressed prices to trigger any death spirals.

Moving Forward

It sucks to have the price of ICP drop so much. I feel that pain. However, I don’t consider a node reward death spiral to be a likely event… it wouldn’t even make my top 5 Internet Computer Worries list at this time. That could change over time, but for now I don’t lose sleep on this. I’d recommend the same for you.

ICP Analysis NFTs

Just a quick reminder that if you like this content, consider buying an ICP Analysis NFT (at CrowdFundNFT). Here is more info if you are interested.