RE: 20% APR For Hive Dollars | Decentralization In Action
You are viewing a single comment's thread:
Increased interest rates don't come for free. It's at the cost of Hive inflation and inflation related risks. At the same time we rely on the witnesses as stewards of this setting, to be responsive to market changes. While the overall rate of HBD savings is low, such interest rates can be sustained, but we could easily and quite quickly reach a savings rate where 20% interest means a substantial increase or risk related to Hive inflation.
Another case where we rely on witnesses as stewards of a consensus setting is the price of Hive accounts. When demand for Hive accounts rose earlier this year, along with the cost of accounts due to the price of Hive, witnesses allowed no changes to the base price of a Hive account to allow for the continuation of onboarding of a large number of users.
I am not totally confident that witnesses will be responsive enough to bring down interest rates when they need to in the future.
Almost 2M Hive was burned as the result of this. The 99.5% of this was Splinterlands who were getting $10 (I believe) selling spell books, so they were able to easily eat that cost while profitably providing new accounts. This massive burst of new Splinterlands users was handled with free account credits for a while. The 15,000 users/day were mostly bots and settled down to much less now after Splinterlands made some changes to make it less advantagous to game the system with thousands of accounts.
The significant spike over $3 was very short lived. As a witness it is in the best interest of everyone to not make rash decisions based on potentially short term issues.
Onboarding non-Splinterlands users was mostly handled with free account claim credits and were mostly unaffected.
That being said, ~$3 for an account (which costs the blockchain resources for eternity) is not unreasonable. Most legtimate users will easily make that back on their welcome post (something that was extremely unlikely in the past, I made $0.03 on mine for example).
I personally considered making changes to the account creation fee, but it wasn't a high priority to me or anyone else. Especially with how resource intensive new accounts are. It's also fair to say 99% of the new accounts were basically spam.
I agree with most of this except that I think the free account credits are facilitating much of the new account spam. Were it not for that, a fee of 1 HIVE or such would likely be sufficient to be a minimal barrer to legitimate accounts while deterring most spam. I have no idea, however, about the long term resource usage. It's definitely an issue. I think we'll hear more from @blocktrades at some point when they report on their review of RC calculations. They've already said that new accounts are a very expensive operation.
Most of these accounts were paying $10 for a Splinterlands spell book (requirement for all new splinterlands players) as they were bots used to play the game to farm rewards. So there was a barrier to entry for the end user and for Splinterlands when the free credits dried up.
I'm not really sure why no one wanted to change the account fee, but there has always been more discussion and recognition of the need to adjust the APR (or at least for a long time; I think back in 2016-2017 the account fee was adjusted more regularly)
It is true interest payments ultimately come out of overall Hive inflation and dilute stakeholder vests. The same is true for all Hive payments & rewards: witness rewards, content rewards, HDF, etc. When we look a the numbers we can see HBD interest payment for the month of March was only $26,500 HBD. If nothing changes and approximately this much is paid out every month, this will cost Hive about $300k HBD a year.
By comparison, Hive distributed more than $10 million in content rewards in 2021. If you add witness rewards and dhf funding these numbers go even higher. Probably would add another $3-4 million.
All of these dilute Hive investor holdings. However, we still invest because we know this money is well spent and will lead to appreciation of the value of the underlying asset Hive.
It is not easy to tell how 20% APR would change things. We can't tell until we try. If this doesn't attract more investors who would put in their money into HBD, nothing changes, no harm done.
If all of the sudden a lot of big money start flowing in, the price of HBD will go up because supply is low. If HBD price goes up, hbdstabilizer will capitalize on this opportunity and make millions for DHF.
Another option for HBD investors will be to buy Hive and convert to HBD, this will create massive demand for Hive which should lead to price appreciations.
In both cases, I think stakeholders will be happy.
Regarding the account creation fees, I think this fee should be in HBD, and not in Hive. This way the fee will always be predictable, and parameter won't have to be adjusted every time Hive prices change significantly.
The difference between the inflation caused by post payouts, witness rewards, staking rewards and the DHF versus inflation caused by HBD interest rate, is that the former are all subservient to the base rate of Hive issuance. They all combine to form the "7% per annum" or whatever the rate is now. HBD interest is an entirely separate source of inflation unbound by the rest of the consensus rules around inflation. HBD interest creates inflation at a rate which can change based on market factors and witness decisions, it can increase or decrease based on changing demand for HBD savings (which can also happen suddenly, there is no theoretical limit on how quickly the $2.8 million UpBit HBD could be put into savings) and changing witness policies.
Under current conditions, with sufficiently low savings rates, inflation from HBD is still low, but that can change quickly, and the compounding nature of HBD savings interest is likely to increase HBD savings rates over time. I do think 20% interest as things stand is not likely to substantially increase inflation in the short term, my concern is if it will be reversed when it may happen in the future.
Particularly I am concerned because there is a lot of ponzi-ish thinking in the crypto space, including in Hive. The supply of HBD is effectively a loan being taken on by the entire Hive community (both liquid and staked holders). We are currently getting a nice benefit that the vast majority of HBD has 0% interest, which effectively means HBD speculators are providing a loan for free. Since USD depreciates in value over time, it can be fair to think of it as a negative interest rate loan (HBD speculators are paying us to take their money).
The economics of 20% or higher interest are very different. To justify taking such a loan, the money loaned must provide value that matches a 20% gain over a year. A company can justify that if it can be invested into producing products which generate sufficient value (eventually revenue from selling the products) to justify the cost. The justification for Hive stakeholders paying that rate is much less clear - in principle growth in the network increases the value of the network, and perhaps the underlying token - but the mechanism by which taking on a 20% interest rate loan gets us that growth has not strongly been laid out. Comparing us to the corporate bond market, the index of junk bond yields (ICE BofA US High Yield Index) rarely go above 10%, currently only 4%. This is what companies rated as "junk" are paying on loans.
By paying more than the rate of inflation, we are subsidizing the growth and network effect of what could itself be a valuable product if it is more widely adopted. But the higher a rate we pay for that growth, the more we are moving into Ponzi-ish territory where we can't actually justify the rate we are paying.
Sure, but even if that happened the inflation from HBD savings would still only be around 0.2%, or $700K out of $10-13 million overall. This just isn't that significant.
You would have to see a dramatic change in supply of HBD relative to HIVE and the proportion of HBD in savings for the interest to really become significant. At that point (or along the way) witnesses should (and I think would) reassess.
Also, I think witnesses might reassess if exchanges started using savings without paying it out to customers. One of the reasons to only pay interest on savings was to support higher APR while reducing the windfall to exchanges
700K out of 13 mill is 5%, it's not earth shattering but it's not trivial either. But I see your point, it's unlikely to happen at a rate that witnesses would not respond to.
I meant the contribution of the interest inflation to overall Hive inflation would be 0.2%.
$700K over $380M market cap.
I understand your concerns. However, we don't even know how much this HBD interest will be utilized. We need more data to evaluate it more. We can't get this data without trying.
I agree with this. It likely wasn't originally because SBD didn't 'activate' until after the first post payouts (3-4 months after chain launch). And then SBD/HBD didn't work very well. So there was never motivation to do it. It also doesn't seem like it is ever going to be a high priority change for development, but I won't say it can't happen.