On Cryptoasset Monetary Inflation

Every week the Mosaic research team will delve into important topics within the cryptoasset space.

Mosaic

Published in

Mosaic Blog

18 min readAug 16, 2018

Lanre Ige

Introduction

There has been an ever-ending debate about the legitimacy of monetary inflation¹ for different cryptoassets, in today’s article, we will take a closer look at monetary inflation to help demystify the argued pros and cons of various inflation regimes.

The first part section of the article outlines the technical and economic reasons for inflation in cryptoasset networks like Bitcoin, Ethereum, Zcash, LivePeer, and MakerDAO.
Next, we consider the problem of solving for the correct amount of inflation for a given cryptoasset; where inflation correctness is defined as the amount of inflation which optimizes (for a particular goal) the actions of the agents receiving the newly minted cryptoasset, whilst protecting the interests of all existing participants in the ecosystem.
Finally, we consider the various arguments for and against inflation and square them with earlier findings.

Justifying monetary inflation

Bitcoin

Many cryptoasset networks use inflation to mint new tokens and reward certain users for completing specific tasks. For example, in the Bitcoin network miners are incentivized to mine blocks due to the potential of them earning transaction fees from users and the block reward from the block they’ve mined. The Bitcoin block reward currently sits at 12.5 BTC and is on pace to halve in May 2020 (the estimated date fluctuates) to 6.25 BTC. The number of bitcoins generated per block decreases geometrically, with a 50% reduction in bitcoin creation every 210,000 blocks, or approximately every four years.

For comparison, the cumulative transaction fees a miner would expect to receive from a mined block (06.08.18) is around 0.146 BTC. Most of a miner’s incentive to mine a block is currently tied to the block reward rather than transaction fees, but this will change due to Bitcoin’s finite supply of 21 million. The number of bitcoins generated per block decreases geometrically, with a 50% reduction in bitcoin creation every 210,000 blocks, or approximately every four years. As a result, the number of bitcoins ever in existence will not exceed 21 million. By 2140, all things remaining equal, block rewards will have ceased and miners will be entirely compensated by transaction fees.

Ethereum

At a high level, inflation on Ethereum is to some extent similar to Bitcoin; for a valid block successfully appended to the blockchain, a miner is rewarded with 3 ether (ETH). Moreover, miners who include uncles — valid blocks that were constructed and rejected by the network are still rewarded for producing them, unlike with an orphan block in the Bitcoin network where miners are not rewarded for producing them. Miners who produce uncles included in a block formed by the successful miner receive ⅞ (2.625 ETH) of the block reward with a maximum of 2 uncles allowed per block; whilst ⅛ of the block reward is allocated to the successful miner. In addition, miners receive transaction fees based off the cumulative Ether paid by users to carry out various operations on the Ethereum Virtual Machine (EVM) — each with corresponding Gas expenditures — and a given gas price.

Two key differences between Ethereum and Bitcoin monetary policy are:

(1) the former does not have a fixed supply — Ethereum does not currently have an upper bound (though relatively predictable) on Ether supply; and

(2) Ethereum’s transition to a Proof-of-Stake (PoS) algorithm will likely affect its monetary policy.

Zcash

Zcash’s inflation schedule is identical to that of Bitcoin — a total supply of 21 million with a block reward that halves approximately every four years (ie.840,000 blocks). One of the key differences between the inflation in Bitcoin and Zcash is who the block reward goes to. At first, 50 ZEC will be created every ten minutes (12.5 ZEC with a blocktime of 2 minutes 30 seconds), with 80% of the newly created ZEC going to the miners, and 20% ZEC to the Zcash Company stakeholders — founders, employees, advisors and investors.

Overall,10% of total ZEC 21-million maximum supply will go towards the Founders’ Reward.

The Zcash team has implemented a Founders’ Reward to support ongoing development and to start a non-profit organization, the Zcash Foundation, self-funded by the Zcash core team contributions. By 2020, after the first four years the ZEC created will drop to 25 ZEC (halving, just like in Bitcoin), while 100% of the block reward will be disbursed to miners alongside transactions fees.

Livepeer

The Livepeer Token (LPT) is used to incentivize participants in the network. Its main role is as a staking token that participants — who want to contribute their processing power and bandwidth in the service of encoding and distributing video — stake in order to coordinate work on the network and ensure that work is done correctly and reliably.

To quote from the Livepeer’s white paper:

It serves as a bonding mechanism in a delegated Proof-of-Stake system, in which stake is delegated towards transcoders (or validators) who participate in the protocol to transcode video and validate work. The token, and potential slashing that occurs due to protocol violation, is necessary in order to secure the network against a number of attacks.”

The Inflation Rate of LPT is programmatically set to be proportional to the total stake of the network’s participants and therefore helps incentivize active participation. Thus, participants who constantly stake and perform ‘correct’ work are rewarded in line with the general inflation rate due to their proportional share of the total bond staked. The inflation rate of LPT token’s inflation rate is a function of the participation rate on the Livepeer network (how many users are staking LPT). The initial supply of LPT was 10,000,000 and currently sits at around 10,270,13 (as of August 7, 2018).

MakerDAO

MakerDAO (MKR) is the governance token for the Dai stablecoin system — an on-chain collateral backed model. The system relies on Collateral Debt Positions (CDPs) which are smart contracts containing other cryptoassets such as Ether (technically Pooled Ether, Ether which is contained in a ERC-20 smart contract wrapping) in order to create Dai. In other words, one must use Ether as a collateral to create and eventually get the proportional amount of the Dai stablecoin. Once the Multi-Collateral Dai version is launched, the MKR token — acting as the recapitalization resource — will be inflated and sold off when necessary to support the stablecoin system. The current supply of MKR is 1,000,000.

Inflation-level optimization

The four cryptoassets discussed above were chosen because they all use inflation for (at least) one of the following reasons:

(1) efficiently distributing the protocol’s coin or token; (Bitcoin and Ethereum, Liverpeer)
(2) incentivizing correct deed and participation on the network; (Bitcoin and Ethereum, Liverpeer)
(3) funding early stage investors and contributors of the protocol (Zcash);
(4) maintaining the security of the network.(Bitcoin, Ethereum, Zcash, MakerDAO)

Bitcoin and Ethereum have inflation regimes that fall under (1), (2) and (4). Liverpeer’s inflation regime falls under (1) and (2), MakerDAO falls solely under (4), whilst Zcash’s inflation regime falls under (1), (2), (3), and (4).

Inflation for distribution

Block rewards are used to distribute coins and tokens to the network, as opposed to distributing tokens through a crowdsale or airdrop. Assuming the set of miners is sufficiently decentralized, it can be argued that this is a good and efficient way to distribute the coins of a given network.

Inflation for blockchain security

Networks like Bitcoin and Ethereum reward miners for validating blocks (and therefore transactions) using capital to secure the respective networks through block rewards. The justification for using block rewards to reward miners for their work as opposed to simply transaction fees — as part of the case for blockchain security — is only briefly mentioned in the Bitcoin white paper. Satoshi writes: “The incentive [of the block reward] may help encourage nodes to stay honest. If a greedy attacker is able to assemble more CPU power than all the honest nodes, he would have to choose between using it to defraud people by stealing back his payments, or using it to generate new coins. He ought to find it more profitable to play by the rules, such rules that favour him with more new coins than everyone else combined, than to undermine the system and the validity of his own wealth.”

This method can be argued to be a way to greater align the incentives of agents (the miners) as to secure the network according to the underlying purpose of said network. As miners accrue more coins it increases their incentive to act in such a way to ensure that their coins maintain their value — such as by continuing to mine in an ‘honest fashion’, especially if they believe the coins will be worth more in the future. Miners are incentivized to work on valid blocks, as invalid ones will be rejected by the network and their mining rewards will then not exist in the longest chain. This point is further reiterated through the 100-block cooldown period for block reward coinbase transactions in Bitcoin. There are also theoretical concerns which may further explain why block rewards are intrinsically important from a blockchain security perspective. This point will be expanded upon later in this article.

Inflation for participation correctness

Certain networks such as Livepeer incentivize actors who stake native tokens and perform actions on the network (such as transcoding video). The absolute amount of the inflated tokens relates to the amount of tokens that these actors stake and the correctness of their actions.

Inflation for protocol-level development incentivization

As previously mentioned, 10% of block rewards on the Zcash network is the Founders Reward.

The general cases for inflation

Redistribution and incentivization

A common argument made by proponents of inflationary monetary policy is the ability of block rewards to help prevent wealth concentration in the cryptoasset network. Individuals who may choose to mine many years after the launch of the network still have the ability to receive newly mined coins. Also, an inflationary monetary policy can be argued to disincentivizing users from hoarding and encourage them to instead engaging in the network (through deploying and engaging with smart contracts, for example).The question of whether or not hoarding is an action a given network wishes to disincentivize depends on its stated function — for example, Bitcoin Maximalists may see no need to disincentivize hoarding. However, this redistribution argument is predicated on the assumption that coin issuance does not consistently go to a concentrated elite. If this were the case then inflation would, in fact, make the system more unequal over time as the wealth distribution tends to resembling an oligarchy.

We can view the Zcash founders’ reward (and similar inflation funding schemes) as other examples of redistribution-justified monetary inflation. These schemes use inflation as a way to reward developers and investors who helped facilitate the launch of a given cryptoasset, whilst incentivizing them to maximize the value of the network up until the end of the vesting period. The Zcash founders’ reward has been met with criticism due to concerns of centralization by the Zcash company, which seems diametrically opposed to the proclaimed ethos of a decentralized, privacy-oriented and censorship-resistant cryptocurrency.

Whilst Zcash is a relatively primitive example of inflation funding, Tezos, EOS, & Decred aim to use derivatives variants of inflation as a funding vehicle for projects and proposals decided upon by token holders and other stakeholders.

Other attempts at inflation-funding from Tezos, EOS, & Decred have understandably not been met with the same kind of criticism. The reason for this is likely due to control token holders have over the flow of funds in the latter’s case, whilst in Zcash’s case there is no in-built governance mechanism which would allow users to fork the code and easily remove or make changes to the founders’ reward as with ZClassic.

The arguments against the founders’ reward due to centralization are similar to arguments for block reward reductions due to miner centralization. Generally, the ‘fairness’ of a given inflation-funding mechanism is predicated upon the ease by which network stakeholders can make changes to the mechanism and whether or not the recipients the funding are seen as a centralized set of actors or fairly decentralized group.

Fee-sniping mitigation

There is some evidence — in a paper written by Narayanan et. Al — which suggests that miner incentives will start to go haywire as their rewards shift from block rewards to transaction fees².

The paper’s key insight is that if miners are only paid in transaction fees, the variance of their reward is likely to be very high due to the randomness of the block arrival time. Therefore, it could become very compelling for miners to fork a chain to obtain a “wealthy” block and “steal” the accumulated fees. After a block has been successfully mined, there are few uncommitted transactions available — and hence minimal fees to be collected — for mining the next block. Until this pool of uncommitted transactions builds up again, it is more profitable for miners to capture it by mining the bundle of ordinary transaction fees in the previous block rather than mining on top of it. A miner would not be able to use a similar strategy with the presence of a block reward, as such a strategy would cause a hard fork — likely to be rejected by the rest of the network.

Source: http://randomwalker.info/publications/mining_CCS.pdf

Narayanan’s paper offers a rethinking of the role block rewards and inflation place in the design of a crypotasset network. The prevailing view is typically that block rewards are not necessary from a blockchain security perspective in the long run. These results suggest that block rewards are necessary for crypto asset networks, and monetary inflation is a necessary feature for the effective design of said networks. There has been little analysis done, besides the Narayanan paper, into how mining stability is affected by little or no mining rewards.

Narayanan’s paper is an often used as an example of why fixed supply is a bad idea from a security perspective. However, the fee-sniping problem outlined by Narayanan had already been thought of by Peter Todd in 2013, with several mitigations theorized and implemented by the Bitcoin Core team which don’t involve changing Bitcoin’s monetary policy. For example, the following has already been implemented into Bitcoin:

The Bitcoin pull requests #2340 and #6183 by Peter Todd set nLockTime on wallet transactions such that they can only be mined by the next block, rather than a block orphaning the current best block. This discourages ‘fee sniping’; the miner’s strategy is to deliberately mine blocks that orphan the current best block. However, with nLockTime the miner will soon run out of transactions they could put into their new block, which means they’d need to mine another block on top to claims the fees of other transactions within the mempool. Due to the limited block size, another miner could employ the same fee-sniping strategy the first miner had used to orphan the new best block. The more important reason is to discourage “fee sniping” by deliberately mining blocks that orphan the current best block. This theoretically weakens a miner’s incentive to fee-snipe even in a 0-block reward world.

The latter are potential mitigations:

Ensuring that the blocksize is meaningfully small compared to the transaction load or size of the mempool. This ensures that there is always a sufficient backlog to overcompensate for variance in block arrival times.
Arrange a voluntary fee forwarding scheme for miners in cases where fees are unexpectedly high. Gregory Maxwell mentioned a construction wherein ⅓ to ½ of the difference between income gained for the next block and the next two blocks (including any previous paid-forward payments).

The key takeaway from the fee-sniping problem in a 0-block-reward world is that there are mitigations that have already been implemented by the Bitcoin Core client. Similar mitigations could potentially be implemented into other cryptoasset networks and one should not assume that inflationary monetary policy is the only solution to the problem of fee-sniping in the long-term.

The general cases against inflation

It was important to outline the different kinds of cryptoasset monetary inflation in order to better understand the different arguments for and against it.

The Sound Money Argument

A common argument against monetary inflation in cryptoasset networks like Bitcoin is the ‘sound money’ thesis. Sound money is generally defined as: money that has a purchasing power determined by markets, independent of governments and political parties — such as precious metals like Gold when used as a medium-of-exchange. To quote from ‘The Many Faces of Bitcoin’ [on the topic of sound money advocates]:

“These individuals believe that, for the foreseeable future, the goal of the Bitcoin project isn’t to facilitate the buying of coffee, but to become “high-powered” money, an even better form of gold. They claim it to be a digital asset superior to physical gold due to a truly limited supply and more deflationary emission. They also claim that, if used properly, it is unseizable, unhackable, arbitrarily unprintable, and is an attempt to engineer a superior form of money.”

In terms of monetary policy, this group — often called Bitcoin maximalists — believe that Bitcoin’s identity as sound money comes from its hyper-disinflationary money supply. The Sound Money argument is implicitly (and sometimes explicitly) the core driver behind most discussions around monetary policy for Bitcoin — and to some extent cryptoassets that have since followed. An analysis of this thesis is beyond the scope of this article, but it is important to be able to demarcate between the different arguments used by proponents of different monetary policies.

A side-result of the sound-money and digital-gold theses, which have been a recurring narrative throughout Bitcoin’s existence, is the general sentiment that — all things being the same — scarcity is good (from an investment perspective) for a given cryptoasset. This belief has driven many of the discussions behind removing or reducing the block reward for Ethereum over the last two years. Consider the following EIPs:

EIP-186: Reduce Ether issuance before proof-of-stake — “A reduction in the issuance of Ether (Ether) is very likely to be price-supportive and lead to increasing investments in the platform and to help ward off speculative attacks on the value of Ether by promoters of competing platforms who offer, or plan to offer, reduced token inflation rates.”

EIP-960: Meta: cap total ether supply at ~120 million — “In order to ensure the economic sustainability of the platform under the widest possible variety of circumstances.”

Part of the justification in both EIPs is driven by a belief that a reduced block reward would be ‘price-supportive’ (and improve the ‘economic sustainability [of]) for Ethereum. Such beliefs are driven by Ether (and other crypto asset) investors having an understandable aversion to dilution of their investment in Ether. However, as we’ve shown, monetary inflation has different roles in different cryptoasset networks; if the case for inflation (through the economic value created by security and participation improvement) outways the negative effect of dilution, then the ‘price-supportive’ argument against inflation is weakened.

For cryptoasset networks which do not have a sound money philosophy embedded into their design, committing to supply caps or no monetary inflation does not make sense — prima facie. For example, it would not make sense for networks such as Livepeer or MakerDAO to commit to a fixed supply given that both networks use (or will use) inflation as essential mechanisms to help support their cryptoeconomies. For Ethereum, the question of whether or not inflation is a necessary good or evil from a technical perspective is not entirely clear.

Proof-of-Stake Economic Security

As previously mentioned, the technical and economic arguments for changes (such as the introduction of a supply cap or large block reward reductions) in monetary policy to inflationary monetary supply cryptoasset networks like Ethereum have yet to be properly elucidated. However, two interesting strains of arguments have come from the Ethereum community, both relating to the expected adoption of Proof-of-Stake (Casper):

Decreasing block rewards can help reduce the likelihood of a miner-driven chain split due to the Proof-of-Stake switch.
Under Proof-of-Stake the amount of Ether validators are likely to stake (and thus the cost of a 51% attack) is related to the total value (in USD, CNY, BTC, etc) and therefore an inflationary monetary policy could weaken the security of Ethereum in the long term.

For Ethereum, block rewards are expected to be lower under a Proof-of-Stake regime; PoS advocates argue that due to miners not being required to consume large amounts of electricity and invest in specialist equipment, fewer coins will be needed for issuance in order to motivate validators. Those who believe (1) argue that miners may be incentivized to reject a PoS fork (and also fork away from Ethereum’s current difficulty bomb), therefore reducing block rewards a long time before the switch to PoS could help smoothen out the negative impact on miners. This argument is weak as it assumed that miners would be naive to block-reward-reduction smoothing; instead, miners are more likely to: (1) reject a hard fork any reduction they deem unfair; (2) hard fork away from the mining difficulty bomb when the effect becomes noticeable, or; (3) reject the PoS switch hard fork. As such, (1) is a weak justification for a block reward reduction as it is unlikely to significantly reduce the likelihood of a miner-driven chain split.

(2), in essence, argues that in order to ensure the economic security of PoS Ethereum, Ether must have strong ‘store-of-value’ qualities and therefore a finite supply or extremely low block rewards. The security of PoS relies on a 51% attack on the system being expensive to the point of infeasibility, therefore — for fixed amount of Ether staked — the security of PoS Ethereum increases as the price of Ether increases. It follows from this line of argument that the inflation regime for Proof-of-Stake systems should be designed as to maximize the per unit price of the native cryptoasset.

This argument relies on the assumption that economic security of a Proof-of-Stake network is solely derived from the per unit value of the native cryptoasset and that the only purpose of monetary inflation is to maximize a network’s economic security.

We have shown that, due to potential problems like fee-sniping, a network’s economic security may be affected by monetary inflation, outside of the per unit value argument. Moreover, one cannot discount the potential of using inflation as a tool for redistribution — as is the case with Decred and Zcash; committing to low or zero monetary inflation removes such a possibility.

Wrap up and further thoughts

This article has overviewed a portion of the inflation mechanisms across the cryptoasset space, explaining how they work and why they are designed and needed. Moreover, we covered various arguments for and against monetary inflation in cryptoasset networks. The key insight is that the topic of monetary inflation within cryptonetworks is a multi-faceted and nuanced topic. To date, most discussions around monetary inflation have avoided looking at the topic objectively due to ulterior motives or agendas people may have. Any debate regarding a change to an inflation schedule or a reduction in a block reward on a given network must elucidate: firstly, why the inflation existed in the first place and then next understand how changes may impact the network going forward. We hope this article has helped provide some more clarity on this topic.

There are further problem areas related to inflation which still need to be explored:

In a Proof-of-Work network what is the minimum-viable block reward possible which does not compromise on the cryptoasset network’s security?
How feasible are fee-sniping attacks on networks other than Bitcoin? How do Peter Todd’s proposed fixes affect miners incentives and can they be applied to network’s with varying block sizes like Ethereum?
How well does high inflation disincentivize hoarding/incentivize active participation in the network, especially given the current speculative and return-heavy state of the crypto asset space?

_______

[1] Throughout this article will use the term inflation to refer to monetary inflation, not price inflation.

[2] Peter Todd recognized the same problem in 2013.

End of weekly research report

We hope you have enjoyed reading the Mosaic researchers’ weekly report and stay tuned to the next article. Click here to visit our website.

This article is intended for informational purposes only. The views expressed herein are not and should not be construed as legal or investment advice or recommendations. Recipients of this article should do their own due diligence, considering their specific financial circumstances, investment objectives, and risk tolerance before investing. The individuals contributing to this article have positions in some or all of the assets discussed. This article is neither an offer, nor the solicitation of an offer, to buy or sell any of the assets mentioned herein.