Introduction

Last Thursday, September 15, marked a turning point in the short but rocky history of cryptocurrencies. The second largest protocol in terms of valuation and the first in terms of usage and flows, Ethereum, made an update changing the very nature of its consensus algorithm from proof-of-work to proof-of-stake, and changing its monetary policy in the process. That's closely related to what we'll be talking about in the first part of this analysis: the monetary policy and the layout of the token economy brought by crypto-currencies, more than 12years ago.

Of course, we must never forget that such economics considerations are linked to technical underpinnings in crypto protocols. Nevertheless, this article will focus on the economic and financial aspects.

Indeed, we could ask ourselves the question of what will change in the Ethereum issuance policy and how will it impact the price of the ether. What will the redistribution of fees and rewards to validators imply, how the combination of the new monetary policy with the EIP1559 (1) allows ether to be a potentially deflationary "ultrasound money". We will also reflect on the consequences and risks that could be induced by these changes. But first we need to take a look at the history of the token economy in order to understand the usefulness of such models, which is unfortunately often misunderstood.

In this essay in three parts, we will discuss the history of the token economy and how the different mechanisms came about. This will allow us to deepen the subject by better understanding the different choices of architecture and how the environment became what it is today. We will obviously start with layers 1’s incentives, because this is where everything started with Bitcoin. Then we will gradually move forward in time to talk about the second major era of token economy, the ICOs wave. We will understand how this period marked tokenomics and why some questionable directions were taken. We will finally talk about the near past and present with the revolution brought by DeFi and its protocols, the democratization of DAOs, and the appearance of tokens carrying all the value of the service without any parallel legal entity capturing all or part of this value: what is commonly called real yield. Finally, after having dissected all these different points and innovations, we will try to understand and anticipate the future directions of the token economy.

The Bitcoin era and forks, incentive tokens to maintain the protocol

As everyone knows, in 2008 the seminal paper of Bitcoin (2) was born. In this whitepaper the monetary policy adopted by Bitcoin is already present. Indeed, when we talk about money, the question of its issuance obviously follows and Satoshi solved it on a model similar to that of the discovery and extraction of raw materials (hence the analogy of mining rather than validation). This distribution could have been considered rather curious in the sense that at the moment when demand is the lowest, issuance is the highest (if we assume that adoption creates demand). But it is quite consistent in the sense that the asset becomes increasingly scarce in terms of issuance even if it is diluted a lot at its creation. Finally, Bitcoin really follows the model of a commodity. The best analogy here is gold: thousands years ago, the Egyptians could find gold by dipping their hand in the Nile, while we now have to drill miles deep to extract it.

But beyond the distribution model, we can ask ourselves the question of the utility of such a distribution and the reason of the orientation of the reward towards the miners. There are several reasons for this, first of all the logic of proof-of-work, which rewards those who spend energy for the security of the network. As they spend energy, it is logical to reward them for the work they do. Moreover, proof-of-work mining is a rather elegant, though incomplete, resolution of Sybil's problem (3) , that a user of a computer network can have an almost infinite number of identities. Indeed, to distribute tokens it is necessary to be able to identify the participants. To do this without a central authority controlling identity, few means exist, proof-of-work mining being one of them.

Now that we addressed the issuing policy and the direction of the distribution, we can ask ourselves the crucial question of the usefulness of such a distribution in time. The distribution of Bitcoin allows the network to be completely autonomous and decentralized. It becomes, thanks to distribution, a quasi-autonomous organism where the incentives allow the maintenance of the network by extrinsic actors. (4) This is the magic of Bitcoin: to have succeeded in making sure that there will always be someone for whom the value of a bitcoin is sufficient to launch his computer and try to earn rewards. And what makes Bitcoin valuable are the characteristics of its network, i.e.: high resilience, incorruptibility, permanence, stability of its monetary policy (with a rare number of units) and finally its free and guaranteed access.

Thus, the first model of token economy is above all that of a commodity which rewards the people carrying out proof of work for the network. The unit is also used to access the network by paying a fee that should incentivize miners to keep securing the network even at the end of the issuance (in 2140 (5) ). Regarding the question of fees and post-issuance, it is possible to speculate on many different futures and this is surely one of the biggest issues for Bitcoin if it is to maintain its value over time.

In 2012-2013 and onwards, we saw many copies of Bitcoin with more or less different issuance parameters. It was also during this period that new consensus mechanisms were introduced with the proof-of-stake introduced in a hybrid way with Peercoin or adopted completely with NXT. (6) It is appropriate to recall that this is also the time when the first ICOs appeared.(7) The tokens had then a new function, that of collecting money in exchange of a part of the future tokens. But this was always for layers 1 that were trying to become a new Bitcoin with different features but not for services and dApps as we see now. Before this period, the issuance of new tokens could be done with the help of proof-of-burn, i.e. one had more or less tokens on the new network (the Bitcoin layers 2 of that time like Omni or Counterparty) by burning a number of Bitcoins, which were thus de facto unusable.

Back then, no one cared about the non-productivity of assets and yields were not yet a topic in the crypto world. Bitcoin did not generate any direct return and nobody cared about it, the first proof-of-stake did but it was inflation or minimal amounts due to fees. This is a point that we will develop in the following, but it is important to understand that at that time, the economic function of the tokens was divided in 4 cases:

• Medium of exchange on the protocol;

• Tokens allowing the use of the protocol for a fee;

• Incentive tokens to maintain the security of the protocol;

• Tokens to fund the development of the protocol (ICOs).

At this point, before 2016, all these uses seem to be clearly validated and, in appearance at least, not so flawed. Of course, there were few issues like the fact that miners needed to finance their hardware and therefore caused a bearish tension on the price - but for some this was a feature that allowed to avoid too much money concentration in the hands of validators, unable to hoard everything. There was also a beginning of criticism on the concentration of tokens in too few hands consecutively to an ICO. Finally, there were criticisms when the teams did not make a "fair" launch of the network and practiced what was called "premine" at the time. Funnily enough, these "unfair launches" became the norm a few years later.

The ICO era, fundraising, buyback, utility (or not!)

The most decisive ICO was the Ethereum one, in 2014, which broke many crowdfunding records with more than $18m raised in Bitcoin. A year later on July 30, 2015, Ethereum, the first turing-complete protocol, was born allowing the network to perform a whole bunch of operations, no longer limited to the various features of Bitcoin. With Ethereum, the possibilities of financial programming became unlimited and more importantly, it became possible to create a services on Ethereum as a first layer, removing the need to have its own blockchain protocol as it was the case before.

The year 2015 and 2016 saw two notable projects born via ICO: the infamous TheDAO which wanted to be a large decentralized investment fund to finance this emerging ecosystem. The idea was to have a token to vote for eligible projects and also to share the proceeds among the members of the DAO. We then see a new function of a token which is no longer to secure a network or to be the token of gas (of expenses) but to represent rights of governance and financial rights within a DAO (Decentralized Autonomous Organization (8)).

TheDAO project was a success in its ICO phase, becoming the most funded crowdfunding project in history (over $150m) and accumulating more than 5% of the outstanding ethers. But it was hacked because of a poor drafting of the smart contract governing the DAO. The hack even caused the Ethereum protocol to be split in order to pay back the investors. As a result, there was a hearing two years later at the SEC (9) on the subject of assimilating tokens to securities and the ecosystem was strongly disturbed about the desire to create complex smart contracts. It was also the occasion to notice that the tools allowing to interact with the contracts were too sketchy because very few people had voted for the patch (10), designed to stop the hack. This is also one of the reasons why complex economies and governance around tokens did not take off.

The second project that marked the history of financing on Ethereum was Augur, a decentralized predictive platform project, aiming to allow betting on anything and everything in a decentralized way, which would eventually even allow decentralized insurance. The ICO also worked very well with almost 2% of the outstanding ethers invested in the project, an interesting fact as it was the first large scale project on the network. More interesting, a new token function emerged with the use of tokens as collateral to initiate liquidity in prediction markets. This was a more complex interaction than what was being done at the time and may have been one of the reasons for the two years of development required for the platform. Which again, due to the lack of environment and satisfactory UX, was not very successful and was perhaps the platform with the highest valuation per active user (37 active users on August 8, 2018 for $250m market cap or $6,7M per user). We can see then that this universe is growing and still suffers from very sketchy UX, however the value proposition of funding projects through tokens was confirmed.

The democratization of ICOs was also due to the establishment of standards like the famous ERC20 (11) created in November 2015. And with the excitement of the market from 2017, we saw an avalanche of projects being financed in this way with a lot of non-legitimate projects or even scams. There were many questionable distributions and truly useless tokens. But it was also an opportunity to test new economic mechanisms, both in the issuance with the birth of bonding curves to participate in ICOs, mechanisms of perpetual financing as the one of EOS, Dutch auctions and many other mechanisms to find a more consistent price and a fairer distribution. But one of the shortcomings was that not many projects were capped, and they ended up with crazy values with no demand once the ICO was over. 

Finally those tumultuous times did not only see new mechanics in financing, but also new economic mechanics theoretically giving the underlying token a value. As we said above, the shadow of being a security had started to loom and all projects tried to be as ingenious as possible in order not to give any financial rights or governance to the token holders so as not to be qualified as a security by the Security and Exchange Commission (except for layer 1s which did not reinvent anything and started to turn to proof-of-stake often with a constant inflation not necessarily thinking about the economic implications behind these choices). We then saw several new mechanisms emerged, from the so-called usage tokens: one would need so many tokens to benefit from such an advantage, an exchange would allow to pay less fees by keeping one's token, another would serve as a mandatory currency to access a service (often a bad economic choice because it confronts notions of currency velocity (12)) and finally one of the most classic, that of the buyback and burn of tokens on a certain pro-rata of the company's profits. One of the iconic tokens of this period, using such mechanism is Binance's BNB which, before having its own chain, was used to reduce fees on the platform while being coupled with a regular buyback operated by Binance based on its profits. 

However, it is clear that most of the tokens from this period were completely overtaken by Bitcoin and did not survive after the late 2017 bear market. And this for a very simple reason, most of them when not useless, did not have a relevant distribution, not enough intrinsic value and often rudimentary uses. The UX being still poor at the time, such cocktail was perfect to allow all the commentators to make fun of the vast joke that is the crypto-economy. Finally, it should be added that far too many projects were actually concentrating their value in equity without sharing it with the token holders. Regulations did not help, but the opportunism of the period and the greed of some project holders were formidable. The economic function at the time can be summarized as follows: 

• Membership and rights according to the number of tokens;

• Token gaining value through buyback and burn mechanisms (indirect representation of a minority share of the equity);

• Collateral tokens used to place a stake (outside of layer 1),

• All of the above.

However, after two beautiful years of bear market (2018/2019) a new bright period was going to open for tokenomics. Crypto-currencies were finally going to move up a gear in terms of system dynamics in economic terms. More complex tools and models were going to allow the emergence of a whole new economy more coherent and sustainable on the long term, probably leading to a slow revolution of token as the best proof-of-ownership support that humanity could have. 

But all this will be for the second part of this essay. Stay tuned.

Footnotes

(1) EIP1559 is an Ethereum Improvement Proposal that had been implemented. It gave a base fee price for miner but burn the amount above this base fee price. See : https://ethereum.org/en/developers/docs/gas/#eip-1559
(2) Nakamoto, S. (2008). Bitcoin: A peer-to-peer electronic cash system. Decentralized Business Review, 21260. https://bitcoin.org/bitcoin.pdf
(3) Sybil Problem come from a case study from a woman with DID (Dissociative Identity Disorder), see more https://en.wikipedia.org/wiki/Sybil_attack
(4) Bitcoin is often compared to a living organism, a very good article about his Mycelium nature https://brandonquittem.com/bitcoin-is-the-mycelium-of-money/
(5) 2140 is an approximation. In 2040 almost all Bitcoin will be mined. As the curve is an asymptote and the mining period is around 10 minutes, it’s difficult to have a precise day.
(6) A small article about the genesis of Proof of Stake https://bitcoinist.com/the-evolution-of-proof-of-stake-from-peercoin-and-nxt-to-algorand/
(7) The ICO thread of NXT : https://bitcointalk.org/index.php?topic=303898.msg3253022#msg3253022
(8) Term popularized some time earlier by the creator of the delegated proof of stake Dan Larrimer through the Bitshares project which aimed at issuing synthetic assets and stablecoins from 2014-2015.
(9) See this sec report https://www.sec.gov/litigation/investreport/34-81207.pdf
(10) During the DAO Hack, a code that permit to stop the hacker was quickly released. But to execute it, it was mandatory to have a synchronized full node and to execute it from the Ethereum client that was terrible at that time.
(11) The ERC20 standard is the most famous one. It standardize the token model on top of Ethereum. See https://ethereum.org/en/developers/docs/standards/tokens/erc-20/
(12) The problem of a token that serve to only pay a service is that the seller of the service will want to have agnostic money and will sell the money. So you have only a proxy money that never take value because there is no saving.

About Shift

Shift Capital is a long term, open-ended crypto fund offering a one-way bet on crypto. We are a global team of 7 people covering the US, Europe and Asian market. Our hybrid and capital efficient investment strategy gives our investors an entire exposure to the crypto market, with venture investments into early stage tier-1 deals, or through an actively managed portfolio of liquid blue chip tokens, or via market neutral strategies.