Nick asked me to post this here instead of just PDF form, so here it is.
I write these words in frustration, not because I have given up hope, in which case I would have already sold my stack and written off the loss, but because I still have hope, and a good deal of it. The network is right on the edge of greatness, but the will to perform the decisive action to take it to that stage, to push it over the edge, seems absent. Perhaps the drawn out development process has taken its toll, perhaps the team is simply being cautious, perhaps I am mis-reading emotions, but whatever the case, the will to act must be found NOW, in renewed vigor, lest the momentum be lost. And this action must be taken on an objective analysis of the current state of affairs, no matter how divorced the conclusions might be from the current decisions.
My proposal includes four parts:
massively and suddenly increase rewards, somewhere to around 90% per year, halving every year. Until transaction fees take over, these rewards will have to remain high for some time
lock these rewards for a period of at least 6 months, with a rolling release so as to avoid a single large token dump
Roughly 30% automatic delegation penalty—NOT a minimum fee floor set, but just simply less rewards allocated to the delegator
major marketing campaign to get the word out, after this new rewards program has been implemented
The Current State of Affairs
Let’s look at reality for what it is: the market has priced Harmony as a VC money grab that failed to deliver on its promises. Erroneous of course, but it is how the market has priced harmony nonetheless. Momentum is not on Harmony’s side. OmiseGo has just been added to coinbase after releasing its second layer scaling solution for ethereum, and Matic will soon release its own. Elrond has been aggressively elbowing its way in to an already incredibly crowded space and, while distant, Ethereum 2.0 development progress remains a steadily growing specter in the background.
There are two main problems, as I see it, currently facing harmony, which of course impact each other and can result in either a vicious spiral downwards, or upwards:
a. Not enough validator decentralization
b. Not enough developer usage.
If one of these were to skyrocket, it would drag the other with it. And while there is a lot of focus by the team on the second point, through grants, the first is only being half-heartedly addressed.
How we got here
Last summer excitement and buzz filled the air. New buyers were joining the project in droves, harmony was getting lots of press, the team was energetic, upbeat, bubbly.
Then—the token swap happened. Or didn’t happen. And it got delayed, and delayed, and delayed again. On each of these delays the price dumped. The test net was broken several times with no easy fixes. Relentless token unlocks from seed investors and VC’s inflated the supply to over 5X that circulating at IEO, aggravating many who bought around ieo, resulting in further dumps in both price and interest.
And now that the network has finally been released (achieving in half the time what Ethereum 2.0 has failed to) everything seems muted. Many developers and investors aren’t even aware harmony’s product is finally live–all of this in spite of the fact that the network is fully functional, live, and more or less exactly what eth 2.0 is trying to implement. Instead, most are discussing layer 2 infrastructure on Ethereum and have assumed the battle for the base-layer protocol is over.
Grants are Good, but not enough
Grants are going along the right direction, and I am glad the team is doing them as that is problem (b) listed above. What in particular they should be focusing on, with these grants, is porting over existing eth projects to harmony, instead of tying to build new ones. There is an enormous reservoir of high quality code built for eth that can’t function properly due to its bloat and slowness: focusing on porting these projects over would be by far the most cost-effective use of the grant money, rather than trying to code new projects from scratch, which with all due respect is completely idiotic. I’m not sure which path is being emphasized, so I don’t mean the former sentence as an insult to Harmony per say. But until the network is actually strong enough to stand on its own, people won’t give much credit to this boost in performance: plenty of networks from Tron to EOS are faster than eth—are they respected? No, for the simple answer that they are essentially centralized servers, and have committed to being such for the remainder of their existence.
Why Increasing Validator Slots Alone will NOT Result in the desired Decentralization
It is true that the upcoming increase in validator slots will indeed lower the barrier for validators to enter the network. But which ones actually will? With rewards being so low, the answer is only holders who are committed to the health of the network, and want to stake not because it is profitable for them to do so, but rather because they want their long-standing investment to appreciate in price.
This is a deeply undesirable state of tokenomics. Relying on the good-will of participants as a strategy for the health of the network is folly: financial incentives for their actions MUST be given. Harmony pumped by more than 15% today (June 21st, 2020): you’re telling me that the network is supposed to rely on an ANNUAL 10% reward rate for validators, when the DAILY volatility of the token exceeds 10%? Lol.
Ultimately for decentralization to happen, Binance’s enormous supply must be more evenly distributed. We cannot hold the network’s health hostage to the whims of whoever binance chooses to delegate to. The only way for this to happen, is for new outside buyers who have no affiliation to the project as of yet, to come in in droves and purchase binance’s tokens, and spin up new validators. And…
THE ONLY WAY FOR NEW VALIDATORS TO COME IN AND PURCHASE NEW TOKENS, IS TO MAKE IT PROFITABLE FOR THEM TO DO SO.
The Bottom Line: Profitability
Lots of discussion has been made so far about how to get more validators entering the space. The bottom line is: profitability. YOU HAVE TO MAKE IT PROFITABLE FOR NEW VALIDATORS TO JOIN THE NETWORK, or at least give the hope of it being so. This is really the end all be all of the conversation, and yet I’ve seen very little of it being discussed. So let me re-iterate: the network has to give FINANCIAL INCENTIVE to new validators to join the network, rather than just relying on micro-managing new entries with sales-pitches and goodwill. The potential appreciation of the token’s price, along with a high rewards rate, is sales-pitch enough for harmony to get all the new validators it needs to become secure and decentralized—this is further fleshed out in the psychology section of this letter. But on top of rewarding validators, speculators MUST BE PUNISHED: while boostrapping a network from scratch, anyone buying into the token to trade and not to use in a dapp or meaningful transaction should be running significant risk for doing so.
Inflationary Criticism of New Token Release
There is of course the issue of inflation, which is almost always immediately brought as a counter argument to the outlined proposal: assuming a fixed market-cap, logically what a massive new printing of tokens would do is not give validators more baseline money, but simply dilute the existing value of money. Which is partly true—assuming a fixed market-cap (actually, assuming a fixed staking ratio and fixed market cap, increasing rewards just extremifies the wealth transfer from traders to stakers that already happens in any proof of stake network).
My response to the critique is two-fold: first, all increasing rewards does is increase the natural transfer of wealth that happens from traders to stakers, which happens in any proof of stake protocol. In a nascent, speculative, and not necessarily battle tested protocol, this transfer of wealth needs to be extreme in order to shock holders of the coin into action. If we assume a fixed percentage of tokens staked, say 60%, then those 60% of stakers are becoming wealthier at the expense of the 40% of traders. If we raise the rewards rate from 10% to 90%, that rate of wealth transfer increases, massively. So traders holding the token for short term speculation will be diluted. They will in essence be punished for not contributing to the network.
So in other words, an increase in rewards will increase the percentage of tokens staked. Speculators will not be able to hold their tokens carefree, they will have to stake them. And if they stake them, the total percentage staked will increase drastically.
But secondly, the market cap will not stay the same: it will rise. When combined with the locking period, this proposal will actually lead to a potentially large INCREASE in price. Why? For the simple reason that the circulating supply will be massively reduced, amplifying the price effect of any new dapp onboarding massively. The flip-side to rewarding stakers with new tokens is punishing those who are not locking up their tokens to stake. No one will be able to hold their tokens in their wallet waiting for the next pump to sell, without risking massive dilution in the process. If you hold tokens, it should be because you actually plan to use them for dapps or other transactions, not to speculate. In a proof-of-stake network, unlike proof-of-work, speculator freedom comes at the cost of the network’s health—a very serious cost when a network is attempting to establish itself as legitimate.
One item along this vein—that is important to stress—is that, for better or worse, the market currently prices harmony as a speculative shitcoin. The critique of my proposal, that Harmony is going to squander its reputation as a stable store of value by increasing rewards, is absurd: the reality is, the market doesn’t view Harmony as anything close to a stable store of value. Such a reputation can only be built ONCE the network has already established itself as relevant and usage is high. Until then, game-changing and headline capturing changes must be made.
Counter-Argument: What about Market Forces?
There is another counter argument to what I have said: that currently, the protocol is actually designed to have high rewards—it all depends on the number of tokens staked. Thus, the line of thinking goes, the market will naturally reach an equilibrium of tokens staked in which everyone is happy—if rewards get too low, people will actively unstake.
This is flawed thinking—and it’s been exposed as such by how the network has evolved so far. The fact is, that the majority of current validators WOULD STILL BE STAKING, even if the lower bound of rewards were 5%, maybe even 1%. Why? Partly because the price for undelegating is so low (only a 1.5 day wait). But…mostly because the majority of tokens lie in the hands of investors who are committed to the health of the network, even if it costs them short term money. As such, the market is distorted, and does not reflect a true satisfactory nash equilbirium.
Whatever the minimum amount of rewards is, given the current token distribution, it will always be met. Plain and simple. Too many holders will stake regardless of what they are making in rewards. So in determining a proper rewards rate, we MUST look at the MINIMUM possible rewards, and increase that drastically regardless of whether or not it is met, because it will almost certainly be met.
Psychology becomes Destiny: Why a Massive and Sudden increase in Rewards is Needed
Trump’s election illustrated a key fact: in a crowded field, any press, even negative press, is far far better than NO press. The crypto space may be the most ruthlessly darwinian explosion of businesses in the history of capitalism; as such it is extremely crowded. Even if tomorrow Harmony makes headlines for having “ponzi-scheme tokenomics” (even though the above proposal is absolutely no such thing), that would still be far better than the current state of things, which is market indifference. Some potential new validators would be caught in the hype, buy tokens, join the network, which would snowball into more and more adoption and usage, in turn snowballing into more people buying the tokens to stake and make money.
A few paragraphs before I discussed how, assuming a constant market cap, new stakers won’t make much in dollars/bitcoin due to token dilution. But…if a massive increase in rewards is combined simultaneously with more network usage through dapps, and therefore more token demand, a strong upward spiral in price will occur, where fresh capital is coming in, buying and staking because it thinks there is a good chance it makes 90% in DOLLARS, not just Harmony tokens. That in turn will decentralize the network further, leading to more dapp porting, further increasing the price…
This is where the psychological component comes into play: a large rewards increase results in a strong psychological kick to the network, giving validators and delegators the temporary but ultimately self-fulfilling illusion of making lots of money. Even though it is in theory partly counter-acted by inflation, this component of market psychology results word spreading, further capital influx, more hype…and becomes a self-fulfilling prophecy. Seeing numbers go up on your screen irrationally plays to human psychology and engineers positive momentum for the network out of nothing.
One last point on the psychology aspect: the market is fickle, and gets tired of news quickly. If the rewards increase is pussyfooted, where say only increased to 20%, and then 30%, trying to inch upwards to find the exact right percentage that will generate sufficient market momentum, the momentum may never be generated at all. Sudden shocks are essential for inertial shifts. So, need to go big at the outset, with a large marketing campaign, and not be afraid of “overshooting” the amount, because there really are no downsides to doing so.
Reasoning Behind Specifics of What I Proposed
There is nothing scientific about the proposed numbers: 90% just seems like a good eye-grabbing number, without resulting in too much instability in the network. I personally feel as though offering 100% or over would be something the market glosses over as just being too high (it doesn’t seem to carry the same gravitas to me as a number starting with a ‘9’), and would be unnecessary, as no other protocol is offering anywhere near 90% anyways. But again the specifics here should certainly be amenable, and are mostly meant to offer a baseline for discussion.
Locking is crucial. It commits the supply staked to be out of circulation, reducing circulating supply massively, and counterbalancing the enormous amount of tokens that have been dumped since IEO—a sort of consolation prize to investors who have held through the unlocks. It commits newcomers to the long-term health of the protocol, on-boarding more validators while incentivizing these validators to build on Harmony so their investment pays off. What we currently have, a 1.5 day unlocking time, will lead to immense network instability if/when the token price increases. In fact, increasing this unlocking time in some way should be at the forefront of any agenda.
A delegation penalty is also necessary—otherwise, there is no incentive to run ones own validator. The network is currently experiencing issues with this: and while several proposed solutions are circulating (such as limiting the amount of tokens that can be delegated to one node), these efforts are ultimately futile: validators will simply create more nodes to circumvent this limit. A fundamental cost must be paid for delegation, a significant cost, regardless of who one delegates to. It is the only way to force decentralization and incentivize active rather than passive contribution to the network.
Simultaneous marketing campaign along with rewards increase: this goes without saying, really. Some sort of massive, energetic, vigorous campaign to get the word out about how much money you can make staking ONE, to maximize the impact of these protocol changes: flooding telegram, reddit, twitter, existing members and validators of the crypto sphere, will all be natural given the allure of Harmony’s superior tech and user friendly staking experience.
Nothing is Lost–Yet
I actually believe that the time spent so far in having the rewards be 10% or so was not a bad thing: there were several bugs that had to be ironed out on both ends of both the validators’ side and the development team’s side, and so a “trial” period with low rewards actually makes a lot of sense: since my proposal includes locking rewards, locking the rewards without fully understanding how validators will interact with the protocol and ironing out last minute bugs would be problematic and lead to lots of frustration. But now we are past that: validators understand the game, they understand how to tweak their machines optimally, and so now its time to give them their due.
To conclude, so long as Eth 2.0 is not implemented, so long as Eth alternatives are valued at billions of dollars, the battle is not over. But decisive action must be taken imminently to push harmony into the foreground, and soon.