The following post aims to clear up some misconceptions around token inflation and staking rewards comparing Proof-of-Stake (PoS) to Proof-of-Work (PoW) networks. It also introduces a metric that highlights the amount of money a network is spending on infrastructure. All data used is based on a snapshot* of Cosmos, Tezos, Ethereum, and Bitcoin.

Many projects promise high token-denominated rewards for participating in staking. But that doesn’t necessarily imply that there's a high return for stakers. To understand why this is the case, you need to understand what staking in a PoS network is trying to achieve. A network that is utilizing PoS inflates the token supply to pay rewards to stakers in return for them putting up their tokens as a security deposit, which serves as the collateral to ensure that network maintainers (validators) follow the protocols' rules.

Cost of Network Security

In PoW networks, all rewards in the form of inflated tokens (block rewards) and fees in the network go to network maintainers (miners). As fees account for just 3.12% of the total rewards paid to miners even in Ethereum (compared to 1.56% in Bitcoin and close to 0% in early-stage PoS networks Cosmos and Tezos), the lion’s share of miner revenues is made from block rewards. In Bitcoin, annual fees and block rewards amount to $3.6bn, or 3.93% of the total bitcoin market cap.

Yearly rewards to miners amount to 4.91% of the total ether market capitalization (4.76% in block rewards and 0.15% in fees)

Miners sell the inflated tokens they earn to pay their hardware and electricity bills and keep some margin as profits. Proof-of-Work token holders are implicitly paying for the collective efforts of miners to maintain the permission-less network through the dilution of their holdings. The relative token supply ownership of a bitcoin holder decreases by about 3.7% per year at the moment because of this mechanic.

In PoS, what is paid for network security is a bit more complicated. There are dimensions of risks involved when staking in a PoS network - the loss of tokens due to slashings and liquidity risks due to having to lock up tokens for an extended period of time, and stakers get compensated for those. Then there is the actual cost to operate the infrastructure required to participate in consensus in a PoS network (running a validator). Every holder is able to stake by running the infrastructure or by delegating their tokens to a validator in exchange for a commission rate on the rewards earned from staking.

In the following, I will argue that money spent on network security to infrastructure providers can be deducted from the average weighted commission rate prevalent in a PoS network:

By weighing the commission rate each staking provider on a network (in this analysis Cosmos and Tezos) is charging with the delegated stake they are securing, an average weighted commission rate can be calculated. This metric shows what percentage of the token rewards delegators in the network are collectively willing to forego in order to securely outsource the work of maintaining validation infrastructure to staking providers.

When multiplying this value with the total value of tokens at stake, we receive an estimation of what is spent on providing the infrastructure in a Proof-of-Stake network (assuming an efficient validator/delegator market and accuracy of on-chain pricing data). The remaining rewards in the network are distributed to token holders that stake and compensate them for the risks they take, but they do not end up with infrastructure providers (in contrast to a PoW network, where all rewards go to miners).

In this way, a PoS network allows for a much more resource-efficient, scalable, and inclusive way of maintaining a permission-less blockchain network (I won’t cover attacks and other issues in PoS or PoW in this article). Looking at numbers gathered in two early PoS networks (Cosmos and Tezos) and the two largest PoW networks (Bitcoin and Ethereum), we can see that the money spent on infrastructure using the metric described above is an order of magnitude smaller (around 10% of the rewards instead of 100%) in PoS compared to PoW (provided the PoS networks are still early stage and comparatively small).

In Tezos 88.99% of the total network rewards go to stakers and 11.01% to infrastructure providers (based on the average weighted commission rate at the snapshot date; in Cosmos: 90.26% vs. 9.74%)

Impact on an Individual's Share of the Token Supply

By staking a PoS token, a staker is receiving a part of the rewards, which counters the dilution through the inflation of the token supply. At the same time, staking means spending resources either to operate infrastructure or to outsource this work. In addition, depending on the jurisdiction, taxes on the rewards generated from staking need to be considered. On the other hand, token owners that do not stake are diluted to ensure the network stays secure (this is the status quo in PoW).

Overall, an entity staking will always maintain a higher share of the total token supply than one that simply does nothing. This is the compensation for the liquidity and slashing risk implied in staking. Whether staking will increase relative token supply ownership depends on fee levels, the overall staking ratio in the network and the tax burden. In an environment with “low” staking participation, a delegator’s share of the total token supply will increase even after considering taxes (as e.g. currently the case in Cosmos). Furthermore, as fees are shared with those staking in PoS, there is an additional avenue to generate income through staking in the longer term.

The share of the total token supply of an ether holder is diminishing by 4.55% per year. A Cosmos atom owner can prevent being diluted by staking his tokens. Before taxes his share of the total token supply will have increased by 3.9% instead of decreased by 5.04% (a spread of 8.94%) if he staked his tokens (assuming snapshot data prevails for a year).

Finally, some networks have new tokens entering the supply in other ways, e.g. vesting tokens from the team or other issuance schemes. If these are considered, the effective change in share of the total token supply of both holders and stakers will be impacted further. This is arguably an even more important factor, but out of scope for this post.

Chorus One enables token holders to securely participate in staking (currently on the Cosmos and Loom Network). We are excited to support scalable PoS base layer protocols with our infrastructure and research and aim to build a community of token holders that want to advance and shape the future of decentralized networks with us. Follow us on Twitter or join our Telegram to learn more.

* Data taken from the following sources on the 9th of April 2019:

Coinmarketcap for all price data (BTC, ETH, ATOM, XTZ)
OnChainFX for 24h fee and inflation data on BTC and ETH (and annualized)
Staking Rewards for XTZ and ATOM fee and inflation data
MyTezosBaker API for staking data and commission rates
Stargazer API for staking data and commission rates