Cosmos Hub ATOM Token and the commonly misunderstood staking tokens — Part Three

13 min readJul 14, 2019

This is part three where we look at the ATOM token and general misconceptions around staking tokens. Part one can be found here and part Two can be found here

The ATOM token

I often see a lot of confusion around what the ATOM token is used for, so let me clarify:

The ATOM token is NOT used for all staking / transactions across the entire Cosmos Ecosystem. It is specific to only the Cosmos Hub. The Cosmos hub is one of many hubs / zones within the Cosmos Ecosystem. There are other hubs live today such as IRIS (which has its own token IRIS) and Sentinel due to launch later this month (which has its own token SENT). Each Zone will also have its own token to incentivise validators to secure their zone.
Transactions fees paid for the Cosmos Hub Do Not have to be paid using ATOM, a wide selection of different tokens will be able to be used to pay transaction fees such as BTC, ETH etc. The incentive for staking is that you will receive a proportion of these fees in the various currencies depending on the number of ATOMs staked.
It is NOT a currency, nor your normal token that you invest in and just HODL on your ledger. It is a staking token used to secure the Cosmos Hub. ATOM is hyper inflationary (which rewards those that stake the token to provide security to the Cosmos Hub and punishes those that don’t stake via decrease in value per ATOM via inflation.
The Top 100 Validators which stake the most atoms are selected for validating / creating new transactions

ATOMs are like ASICs, just as ASICs are a piece of capital you need in order to mine POW chains like Bitcoin, ATOMs are a piece of capital that you need in order to stake on the cosmos hub and earn transaction fees going through that hub. If a lot of ASICs are already in use it is very difficult to attack the network and similarly if a lot of ATOMs are staked, then it is very difficult for someone to buy a large portion of the ATOMs to attack the network. You can read the document explaining the token by the team here as well as the video below (time stamped from 44:30) as well as here

Staking Tokens

Staking tokens are very commonly misunderstood by people, they assume its a passive income where they can earn 10–20% for doing nothing but staking their tokens. Rewards are created by minting new tokens via Inflation, this depreciates the asset of each token by increasing total supply of the tokens. To counter the negative effect of inflation, you can stake your tokens to earn a reward which is greater than the inflation increases. If there was 20% inflation and 100% of the tokens were staked, then there would be no rewards. It’s would just be like projects increased their total supply when doing coin swaps such as VEN going to VET where they increased the total supply of the tokens and everyone received the same proportion. Those that do not stake are punished as they are not receiving the % increase in new supply and so their proportion is diluted.

There are many examples of some version of the following: “Earn a 15% yield per annum when you stake on x network!” This is at best misleading and at worst potentially fraudulent depending on the jurisdiction where these claims are being made. It causes token holders to evaluate and hold PoS tokens on a basis that isn’t applicable or relevant. Even worse, using these words incorrectly can lead regulators to draw unnecessary negative conclusions about how to tax and regulate these networks/tokens: “If you are calling it yield then it should be taxed as income…”
Staking rewards — and the possibility of slashing — are a set of incentives that encourage token holders and validators to secure a PoS blockchain. In return, they maintain or grow their relative share of token holdings in the network. Staking creates the “skin in the game” necessary for good behavior such as running nodes in the network and discouraging bad behaviors like failing to remain online or double signing.
Staking rewards do NOT exist to provide an income stream to token holders. Think instead, “by staking I can increase my network participation (ownership if you like) by 0.3% over the following year” or “if I do not stake, my relative participation/ownership in the network will be diluted by 1.5% over the next 12 months”.
The economic rationale for staking a PoS token is not to receive “yield” (it doesn’t exist) but because you believe that by doing so you will be growing your relative interest in the network and also contributing to significant token appreciation.

The above is taken from a great article which can be found here which explains the commonly misunderstood Staking Token and related terms such as Yield and Inflation.

Basic Example of how this works

To see how it works let’s look at a basic example. For simplicity assume there are only 2 Validators, “Validator 1" and “Validator 2" and there is a current total supply of 1000 tokens. 300 tokens are being staked with each Validator, with the validator for each staking 150 tokens and the delegators also staking 150 tokens. 60% of the total supply is staked whilst 40% is not staked.

Again, to keep it simple rather than do the rewards per block i am just going to use the yearly figures. So, if total supply is 1000 and inflation is set at 20% then there will be 200 tokens to be minted over a year to be used for rewards and added to the total supply. So total supply now becomes 1200 and the 200 tokens are distributed according to the diagram below (and using the commission / staking values in the diagram above)

So now that the 200 tokens have been minted the total supply has now increased to 1200 and we can compare how the proportions of supply have changed.

The users that didn’t stake — initially had 40% of the supply, they have been penalized for not staking and now only own 33.33% of the supply. (Note they haven’t had any tokens removed from them it’s because additional tokens have been minted and they haven’t received a proportion of them by not staking.) — a decrease of 6.66%

Delegator using validator charging 20% commission — Initially had 15% of the supply and now has 15.84% of the supply — an increase of 0.84%

Validator charging 20% commission — Initially had 15% of the supply now has 17.5% of the supply — an increase of 2.5%

Validator charging 10% commission — Initially had 15% of the supply now has 17.08% of the supply — an increase of 2.08%

Delegator using validator charging 10% commission — Initially had 15% of the supply and now has 16.26% of the supply — an increase of 1.25%

You can see how the users that don’t stake get penalized by not receiving rewards with the increase of additional supply. The proportion of supply that they lose gets transferred to those that stake.

21 Day unbonding period

To protect against a validator attacking the network and then immediately withdrawing his stake, the Cosmos Hub is enforcing a 21-day unbonding period. During this period, staked Atoms do not receive rewards anymore, but slashing is still possible. This means your Atoms are illiquid for 21-days after you decide to stop staking. You will not be able to trade them on an exchange etc until the 21 days have passed. There are however exchanges now looking at offering services where you keep your ATOM on their exchange, and they stake them for you. This has advantages of being able to day trade etc whilst still earning rewards to counter inflation. The downsides are that they normally charge high commission (30%), plus security wise its not great to have everyone leaving their tokens on an exchange as has been proven time and time again. The other potential issue is that it gives the exchanges a lot of voting power over the network if everyone uses them which creates centralisation and may be more inclined to vote on for governance that benefits them. EOS has this issue.

Whale Exchange, Newdex, Hufu, Bigone, and several other exchanges and wallets, have been elected as the top 10 BPs. In the meanwhile, the original supernodes, EOS Newyork, EOS42, EOS Authority, and EOS Canada, all have dropped out the top 21 rankings. Huobi Pool continues to see its dominance.
At present, two of the top 5 ranked super nodes belong to one entity: EOSLaomao and Bigone exchanges both belong to the individual Laomao and team. The interests of the two are closely tied, and the strong essentially becomes stronger.
And now we are seeing a phenomenon where an overarching number of top BPs are coming from mainland China, and in other words, we are seeing EOS even more centralized than before.
Most of the top supernodes currently as of publishing date are either based in China or ran by a Chinese team
Brian, the head of the EOS Amsterdam community, also believes that the exchange is considered to be the “leader in the ecosystem”. He is more worried, however, that the supernodes are almost occupied by mainland China nodes, leading to network security vulnerability, centralization and long-term negative PR. This would subsequently bring down the price of the token.

The Rising Trend of Exchanges Participating as EOS BPs; EOS Becoming Even More Centralized

The rise of the staking economy has driven the business of PoS mining to the exchanges, and subsequently altering the…

globalcoinresearch.com

Slashing

Staking is not without its risks and it’s important to choose a secure and trusted validator or risk having your tokens that are staked slashed. On the 29th June the first validator had all tokens that were staked with them slashed by 5% due to a misconfiguration which caused them to double sign a block. Whilst in this case, the slashing was neither the consequence of an attack on the network nor the result of a compromised validator key, it demonstrates that slashing is real and that validators should carefully design their infrastructure to mitigate the risk of losing their own and their delegators’ funds.

How to choose which Validator to delegate to?

The first metric I look at when evaluating validators is how much self-bond they have. If they have 30% or higher self-bond, this gives me confidence that they don’t want to get slashed as much as I (delegator) don’t want to get slashed. When a validator has low self-bond (1% or less), it makes me less likely to bond to them because they are playing with other peoples’ money, and there’s less incentive for them to bolster their setups.
Many of the top validators are highly visible by making their contributions to the ecosystem known. A lot of them have built useful tools that add to the richness of the Cosmos ecosystem, and thus you recognize their brand through their contribution. For example, you would know about a validator because you’ve used their block explorer. All this of course isn’t telling of the hardness of their setups. This part is hard to verify yourself without going into their data centers and auditing their servers yourself. For now, doing your research on what they’ve got set up as described by their website/content is the best option to understanding what kind of setup they’ve built.

https://medium.com/@huobiwallet/cosmos-ama-on-huobi-wallet-d6b75f6ed492

Tendermint uses Proof of Stake where all validators are known before hand. The current maximum amount of Validators is 100. Validators run a full node for the Cosmos hub and provide its security, as well as being able to vote on Governance about future decisions for the Hub. The 100 Validators which stake the most ATOMs are selected. Currently the minimum amount of ATOM staked to be in the top 100 is 39,047 ATOMs.

The amount of ATOMs staked by a Validator is a combination of ATOM’s that the validator personally holds as well as Delegators, those that rather than run a validator, delegate their stake to another validator and receive a % of their rewards depending on the amount they delegate. There is normally a commision fee that the validator takes as a fee as a % of the rewards received for delegating to them which can normally ranges from 0% to 30% (can see in the picture below). This pays for the equipment, wages etc needed to run a secure validator.

Tendermint requires 2/3 of votes for consensus to be reached. Currently 2/3 of the vote are controlled by the Top 16 Validators (so effectively if these all agree to vote on a proposal then that would be sufficient without the input of the other 84).

If a validator / group of validators control more than 1/3 of the vote then whilst they can’t force any changes through, they can prevent any further proposals from being accepted that they don’t agree with regardless of what other validators vote. So the idea is to have the voting power distributed widely throughout the top 100 for more decentralisation.

Calculating the values for Cosmos

Current Total Supply:

There is no fixed total supply of ATOMs and the total supply will increase each year by between 7% and 20% due to inflation.

https://stargate.cosmos.network/staking/pool

{
  "not_bonded_tokens": "71341288426570",
  "bonded_tokens": "170079253911157"
}

Bonded Tokens + Not Bonded Tokens = Total Supply.

The values in the API include 6 decimal places so you need to divide the number by 1,000,000. So to work out the total supply it would be:

(71341288426570 + 170079253911157) / 1,000,000 = 241,420,542.337727‬ ATOM

You then have a minimum of 7% and a maximum of 20% inflation per year on top of that depending on how much has been staked.

Circulating Supply:

The only tokens that are under a vesting period are for All in Bits Inc (AiB, the company doing business as “Tendermint”). They have a total of 23,619,895.81 ATOMs vested which are split into two sets, each subject to a different form of vesting.

The first set consists of 1,777,707 ATOMs allocated to 44 addresses owned by AiB founders, contractors, and employees, current and past. These atoms are non-transferable for 12 months, but can be used for staking and governance. These will become unlocked on the 13th March 2020.

The remaining set of AiBs atoms are held in an AiB multisig and vest continuously starting 2 months after genesis. This is a total of 21,842,188.81 ATOMs.

Each month 992,826.76 of these are released on the 13th (Starting May 13th 2019 and finishing on March 13th 2021.

So Circulating Supply = Total Supply — (Amount Vested by AIB)

Circulating Supply = 241,039,982.546951 — (1,777,707– (21,842,188.81 — (3 x 992,826.76)) (represents 3 months which have been released so far)

Circulating Supply = 220,398,567.016951 ATOMs

Current Market Cap: $872,778,325.39

How to work out Profit from Staking

The Annual reward yield is currently 10.19 % which can be seen from sites such as here

This is the bit where people get confused with staking. They see 10.19 % reward and think easy money, passive income etc. What you need to understand is that these rewards are from new ATOMs being minted and added to the supply via inflation. And so with a higher supply the value of each ATOM is worth less.

Calculate effective reward rate in ATOMs

((100% — Commission Rate%) * Yield Rate) — Inflation

So if you delegate with a validator which charges 20% commission

It would be (0.8 * 10.19) — 7.66 = 0.492% a year in ATOMs

Calculate effective profit in FIAT terms

ATOMs hasn’t been trading for a full year but if we take the first value in CMC which is $6.44 and is currently $3.95 which is a decrease of 38.66% per ATOM. The yearly reward yield is 10.19% so in profit terms its 10.19–38.66 = -28.47%

Profit in USD Terms = 10.19–38.66 = a loss of 28.47% in USD

Calculate effective profit in BTC terms

At the start of trading each ATOM was worth 0.00164490 satoshis, as of the time of this writing they are now 0.00037155 satoshis which is a decrease of 77.41%

Profit in BTC terms = Reward Rate + Change in Price per ATOM in BTC over year

Profit in BTC Terms = 10.19–77.41 = a loss of 67.22 % in BTC

Note that these calculations do not include transaction costs for traffic going through the Hub. Once IBC is released (minimum viable product version is supposed to be at the end of this year, so i would estimate mid next year for full feature version to be released), adoption of the ecosystem will increase and zones will be transferring between each other over hubs then additional revenue is earned via transaction fees of other tokens.

This site you can see the correct value for Total Supply, % Bonded and Inflation Rate https://www.mintscan.io/

Be warned there are some other sites such as https://stakingrewards.com/asset/atom which show incorrect values (for example they say the staking ratio is currently 88.06% which is incorrect and skews the figures for rewards. Mintscan is accurate and the API site that i listed before is direct from the Cosmos Official website so is correct.

Sources for research / Useful Links

As always do your own research, but hopefully this will give a better understanding of the project, and you can see some of the sources i used whilst researching below so you can find out more for each.

Intro to Cosmos, Whitepaper, Token Distribution Details, Diagram showing how Tendermint works, Tendermint details, List of Projects using Tendermint, Cosmos Validators FAQ, Peg Zones, Ethermint, IBC, IBC BI-Weekly Calls, Cosmos SDK, Relayers, ATOM Tokenomics