Maybe they do maybe they don't.
Timelocking stake is a very important core feature of Hive. Delegated proof-of-stake essentially depends on this feature. Without the timelock mechanics in place our governance would be much more prone to attack. Forcing voters to lock their assets puts them into a position where they have skin in the game and something to lose if things go sideways.
Is 13 weeks too long?
That's up for debate, but I've personally never had a problem with it. HiveEngine tokens prototyped a schedule of 4 weeks, which also seems to work, but I would argue isn't as secure when synergized with account recovery and possible theft. Account recovery is a very nice facet of this network, and it's good to have more time to react to it. That being said it is also possible to enact recovery within 24 hours of a hack, so perhaps 4 weeks is just fine in the grand scheme of things.
(less supply) == (more value)
Another argument in favor of a 13 week locking period is that it "creates value" because the more tokens that are locked the less tokens are up for sale on exchanges. It makes sense in theory but my bet would be that this logic is unequivocally false. The reason for this is that it completely ignores the other side of the equation.
A longer locking period means that users are less incentivized to engage with the contract in the first place. There's a reason Hive's unlock period was changed from over a year to 13 weeks. Lowering it had value. Should we lower it even more? It's debatable but I'd argue that it's fine right where it is.
The point here is that there is an unknown "sweet spot" that is the perfect amount of time to lock stake. Not too short to threaten governance or leave idle tokens floating around, and not too long to incentivize exiting or never entering the contract in the first place; The Goldilocks balancing act. Most completely ignore this nuance when speculating on timelocks, which is very poor planning indeed.
Elasticity?
Things get even more complicated when we realize that the Goldilocks number is a volatile function of time and yield. Users are less likely to stake during certain times like a bear market, while they're more likely to stake if implied yields are higher. Higher rewards or even just positive market sentiment result in more risk taking. With this in mind we can think of the sweet spot for timelocking funds to be more like a sine wave that goes up and down just like the price of the market itself.
Should we try to manipulate our financial policy to better mirror what's going on here? Might be a good idea if we actually had a guess where the sweet spot actually was during any given time. However, I'm not seeing a lot of good metrics out there that would help us achieve such speculation, and changing economic policy on a whim creates very bad optics in crypto where there are so many scams and rug-pulls prevalent.
HBD has no power over governance.
When HBD is "powered up" it only gets locked for 3 days and currently earns 20% APR, which is more than HP curation. As Hive spot price continues to crab walk on the bottom many avid users of the network get in their mind that this is "unfair". Weird how it wasn't unfair before number went down, but I digress.
The natural progression of this argument is that 20% is creating too much inflation and this is what's making number go down. Not only is this incorrect, it is dangerous thinking because even if correct this is the absolute worst time to rug-pull HBD yields. Imagine yoinking it right now to 0%. No more extra debt, that would be great, right? Again, this line of logic completely ignores the demand side of the equation. Yes, less supply will be created over time, but demand drops to zero instantly. Millions of HBD would get dumped on the market and Hive would crash down to 10 cents quite fast.
HBD bonds.
The next argument in the progression of 20% yield being unfair or unsustainable is the potential bonding system. Something like, "Well, obviously 20% is too high for a 3 day lock if we implement bonds, because bonds would be a longer timelock and we can't allow those contracts to earn much more (if any) than 20%." This is the main reason I decided to write this post today, as it is a logical fallacy called begging the question.
The assumption being made is that longer timelocks on HBD have intrinsic value in the first place. They do not. Why would they? Because reasons? No one has presented any evidence whatsoever to support such a theory. Why would the sweet spot for HBD be longer than 3 days? One of the best use cases of HBD is that it's more liquid than Hive by design. Taking that away doesn't increase the value, it actually makes it worse. It doesn't even matter if I'm right or not because the possibility of this statement being true isn't even being considered right now.
All that matters is how much HBD exists at a given time.
More HBD in circulation means Hive is being burned to mint it during times of great demand, or at least that HBD isn't being burned to mint more Hive collateral. Can anyone guarantee me that implementing longer timelocks is going to increase the amount of HBD in existence? What happens if and when we do it and the market simply decides to do the opposite of what we want? Why do we act as though investors love to be punished with more risk/commitment and that enacting such punishment will surely make number go up? It's silly, and once again completely ignores the demand side of the equation.
What is a bond?
A bond is just a reverse-loan. Instead of the institution giving the citizen a loan: the citizen gives the institution a loan. The secondary market allows the citizen to sell their contract on the open market before it has matured (before the institution has to pay up). Many are making the argument that building a bond market for HBD and then a secondary market on top of that is a great way to build value. The more I try to game-theory these concepts the less sense they make.
Secondary market
The secondary market is an evolution to the bonding system. It was built as an afterthought to solve the problem of all these illiquid assets floating around. Why would we copy this model on purpose? It makes no sense. We have the opportunity to build a system correctly the first time. We don't have to build a broken bonding system and then tack on a secondary market to fix the liquidity issue because none of it even exists yet.
Not only that, Hive itself isn't even a debt-based system, but rather collateral based. If a bond is a loan, then why does Hive need a loan? Hive, as a network, doesn't need to ask for money. Hive prints its own money by design, further solidifying that a bonding system lacks critical utility of the systems we are talking about copying.
Thought experiment on Hive bonds.
Let's say we add a one month and six month timelock to HBD.
| 3 days | 1 month | 6 months |
|---|---|---|
| 5% | 10% | 20% |
Is this a more "fair" spread?
Who cares?
The real question is if it creates value or not.
I don't think it does.
First of all, there are plenty of people in HBD who were going to timelock their stake for over 6 months anyway, so all of those users are going to stay right where they are at. Best case scenario the demand for debt stays flat. The only difference is that everyone gets punished with a longer staking period. The fact that 20% is locked for 6 months is completely meaningless. This doesn't create value or demand to create a bond. If anything it incentivizes people to leave so their money isn't locked up for that long, reducing demand.
"The secondary market creates liquidity."
Yeah, it doesn't. Creating the bond market is what takes the liquidity away in the first place (for no reason). Let's say users with 6 month bonds want to sell before the contract matures. Now they sell at a slight discount. This pushes up the yield on 6-month bonds to 21%. Now there is absolutely no reason for anyone to buy a 20% yield contract from the Hive network. They buy from the secondary market instead. Again, that's not creating more value. It's a haphazard backwards way of going about things that only makes sense from the perspective of archaic centuries-old institutions within a debt-based economy.
You know what creates even more liquidity than a secondary market? A 3-day timelock, which we already have. What's the disadvantage? Well, anyone can sell HBD whenever they want. The ironic thing about this disadvantage is that nobody really cares about it. If they did they wouldn't suggest rug-pulling yield at the bottom of the market. That's the best way to get people to sell. If anything we should be doing the exact opposite and jacking up yields even higher if we thought that would create an appropriate amount of demand (which it very well may not at this point; diminishing returns).
We should be incentivizing actual liquidity.
Paying people to lock HBD in the savings accounts (lowering active supply) is a silly thing to do in the first place. I've been saying this ever since DEFI 2020. The best allocation of these yields is within an AMM atomic swap contract between the HIVE/HBD pair. That way Hive and HBD would have deep liquidity with each other and we'd actually be getting something out of the deal. However, since none of our core devs seem to agree with this assessment it seems unlikely to ever be built.
Conclusion
For a while I was thinking that bonding contracts and a secondary market to sell them on would be a really great thing for Hive. However, now that I've had months to think about it I just can't understand how making HBD even more illiquid could ever result in a positive outcome.
Bonds exist and cater to centralized leadership. A bond is an investment for both parties. The institution is betting that they can generate more value than the yield they agree to pay down the line to the citizen. A government can use bonds to build infrastructure of fund a war. A corporation can use a bond to fund development of a product. Because Hive is not a centralized institution: bonds make zero sense.
The loan being taken is not controlled by someone looking to create a return. To assume that we will collectively, as a network, generate a return on loans taken at completely random times to be paid back at completely random times is childish at worst and over exuberant at best. The sweet spot is a moving target and requires intelligence to target correctly.
Timelocking stake for longer time periods does not create value. People do not invest in USD bonds to punish themselves. They invest in bonds for the yield and because the government has a high enough reputation to justify the risk of default. USD has utility. If we want to create value for HBD it doesn't need bigger timelocks; it needs bigger utility. We need to focus less on lowering supply and a lot more on creating demand.
These are the tenets of abundance itself.
Are you suggesting that spending time developing bonds on Hive would be wasteful (now)?
Well I mean by definition any decentralized system is going to be redundant and create a lot of "wasted" effort.
Doesn't necessarily mean the thing isn't worth doing.
Yada yada yada I'm just giving my potentially flawed opinion.
The more I game theory the whole bonding thing out the less sense it makes to me.
I tried to get on board with the idea a couple times, but always big questions presented themselves.
Bonds only make sense to me if they are issued intelligently or governed by a free market.
Hive does not need to be taking random loans at random times to pump/dump price randomly.
It has to be way more tactical than that to actually work.
And even if we did create a good system that worked well there is no guarantee it's going to perform better than the service that we already offer (3 day timelock for flat yield). If I was in charge of HBD I'd change the static yields governed by witnesses to variable yields governed by the free market. If I really got my way I scrap it all for liquidity in the form of an AMM internal market. I'll take millions of dollars in liquidity between HIVE/HBD over locking HBD (and removing that liquidity from the market) any day.
That tickles my mind. I would say that the yield is variable since the witnesses can change it at any time, even if in practice there is significant inertia. But, it's probably just a question of semantics.
On the other hand, the idea of having more dynamic yields influenced by the market is interesting. Couldn't it initially be a bit like the witnesses do by regularly publishing the price feed?
Isn't that just what the @hbdstabilizer does?
I feel like you are implying that I've somehow changed my mind (now).
I have not.
The Endless Possibility of Hive Bonds
In this post I mostly just talk about all the reasons why offering a relatively high yield on our debt is fine.
I conclude by saying I need to think and write about the topic more.
HBD Debt Auction
Also gratz on ascending to the #1 witness slot.
Most impressive, sir.
I wasn't implying anything, it was an open question.
Reading your post, my conclusion was that it was not the right time for bond development. I wanted to validate if we were on the same page.
The "(now)" was intended to emphasize that this might be a matter of timing more than of project, that perhaps we should wait for a more opportune moment to do it.
PS: Thank you for your congratulations 😊
I absolutely agree.
Thanks!
Terra with its 0 time lock was a key instigator to its fault and subsequently caused its collapse and its what's saved us.
That's a little different because you're talking about the conversion from debt to collateral and vice versa. That's actually a lot worse than what I'm discussing in terms of potential for systemic failure.
Seems it's a massive burden on the economy in general currently though
Luckily enough, the way I'm reading it is that layer 1 bonds is a non-starter.
I speculate that it would slow down the chain to keep track of all the data.
This is why smteees!™ never happened.
Layer 2 can be just as secure as layer 1, if the devs build it that way.
Then folks can write whatever contracts they want, and we will see what floats and what stincs.
I'm hoping vsc delivers on that, we'll see.
SMTs never happening had nothing to do with blockchain performances. This was simply because Steemit, inc. has never been able to finalize and deliver.
Being secure isn't enough. You also need trust and reliability.
Yeah which is why at best we don't get actual bonds... just different options for how long to lock the HBD for different yields. That would be the same load on the network as we have now.
In that case I am in favor of no change.
Let vsc, or any other team, set the options.
If what I understand of the project is true, then we will have plenty of investment options negating the need for level 1 bonds.
Afaik, the project is being developed as trustless.