Signal Request: Should we increase the SCD debt ceiling?

This has been discussed previously in this thread however the matter became largely unimportant after the Dai supply dropped previously. Several people have expressed that it might be time to renew the discussion considering the recent Dai supply increase and bullishness of Eth. I’ve created a new thread for clarity, both because the previous thread is out of date, and because many of the responses were meta-discussion around polling.


Should we increase the SCD debt ceiling?

Positives

  • Reaching the debt ceiling introduces a confounding factor into the Dai price as demand increases cannot be matched by supply.
  • Reaching the debt ceiling means that would-be CDP users are forced to go elsewhere to borrow Dai against Eth. This reduces income that could be used to burn MKR.
  • Reaching the debt ceiling hinders the natural growth of Dai, and to a limited extent, the DeFi ecosystem.

Negatives

  • Increasing the debt ceiling increases the maximum severity of the peg breaking downwards in the event of a massive printing of Dai.
  • If the system were to be proved to have a smart contract issue, more funds would be at risk.

Please try to think of additional points to add to these lists. I’ll add any I see expressed in the replies to this post.

I have described the process for the progression of this (and future) proposals here. Please feel free to provide feedback, the process is still being developed, and it is open to change.

We’ll start with an initial sentiment poll to determine whether there is a general agreement to increase the SCD debt ceiling from its current value of 100 million. Please use this week to signal and to propose and discuss specific options. If we have a majority (>50%) for change after 1 week has passed, I will add a poll to gather consensus on any specific options proposed.

This initial sentiment poll will remain open throughout the process.

  • Yes, we should increase the SCD debt ceiling.
  • No, we should not increase the SCD debt ceiling.
  • Abstain (I just want to see results)
  • Abstain (I have no opinion)
  • Abstain (I don’t feel I am knowledgeable on the subject)
  • Abstain (I disagree with the poll options)
  • Abstain (I have a different objection)

0 voters

1 Like

I believe I mentioned this in the previous thread, but just to re-iterate: I’d love if we could solve this problem once and not have to worry about it again until MCD launches.

What do people think about agreeing to a statement that empowers predefined ‘action-without-forum-signals’ in the future? Something along the lines of: ‘Until MCD launches, whenever the Dai supply grows to within 10 million of the Debt Ceiling, an on-chain poll will be created to increase the Debt Ceiling by 10 million.’

In general, I think we should be able to agree on future action in-principle as it allows us to increase governance efficiency by avoiding running this process every time we want to do something repetitive.

8 Likes

I initially voted for a 20 million raise to 120 but the +10 option if within 10 million of the ceiling sounds also good.

In general, the true debt ceiling for Eth is likely to be much higher than 100 million, so from a risk perspective it’s seems okay to start raising it incrementally. Tl;dr, debt ceiling is a function of how much risk we can tolerate, given our ability to absorb bad losses. Eth’s propensity to crash is quite low. It’s likelihood to crash straight to 0 is even lower. So for a given amount of Dai outstanding, we’re unlikely to absorb significant losses. I.e., if the debt ceiling is $200, something very unlikely would have to happen for us to absorb a $100 million loss. And even then, MKR could probably absorb it. In short, moving the debt ceiling up in 10 million increments leaving a buffer of 10-20 million is the safe approach IMO.

4 Likes

Thanks for your points here Cyrus.

We’ve had a lot of signalling here, but not a lot of discussion. If anyone feels they have suggestions to raise, or agrees or disagrees, please don’t be shy to share your thoughts.

Personally I don’t feel that demand is going to increase significantly before the end of the year, and that either a one-time increase by 10-20million or an agreement to maintain a buffer of 10 million would be preferable to no action. While I don’t expect demand to skyrocket, we may see it ticking up over 100 million and it would be good to deal with that possibility.

Can I ask why you believe it to be more likely to increase significantly after the end of the year?

I believe it won’t significantly tick up before the year ends, not that it will significantly tick up afterward.

To clarify: I do think demand will pick up after MCD is launched, I have no idea to what extent though. I’m hesitant to say it will be significant, at least not for the first few months or possibly years. Over the last while I think we’ve seen a slow increase in demand, but it’s slow enough that we’ve not yet hit 100 mil. We’re getting there, and it’s worth keeping a buffer, but yeah, I’m not critically concerned.

Ok. Has it been announced that MCD is planned to launch end of year?

No, there have been no announcement of dates. Personally I think it will be before the end of the year based on the recent ramp up of activity. If not before 2020, maybe in January.

1 Like

Righto, it’s been a week and we seem to have a pretty strong consensus for increasing the debt ceiling. Between this thread and the previous one, there were several suggestions for paths forward. This next poll will last for two weeks, and please remember that:

You are able to vote for more than one option in the poll below. Please vote for anything you would vote for in an on-chain poll.

First lets go into detail on the suggested options.


Option 1 - An immediate one-time increase of the SCD debt ceiling by 20 million Dai, to a total of 120 million Dai.

  • Mostly self explanatory.
  • This was suggested in both this and the previous thread and had good support previously.

Option 2 - An immediate one-time increase of the SCD debt ceiling by 50 million Dai, to a total of 150 million Dai.

  • This was suggested in the previous thread and gathered some level of support.
  • This does increase risk more than Option 1.

Option 3 - A commitment to increase the SCD debt ceiling by 10 million whenever the total supply of Dai comes within 10 million of the current debt ceiling. This commitment will last until the launch of MCD or until explicitly revoked by governance.

  • This will solve the problem/risk of hitting the debt ceiling until the launch of MCD.
  • This option requires action(s) by the Interim Governance Facilitator (or the Interim Risk Team?) at unspecified future times.
  • There is some complexity as to the interpretation of ‘when the dai supply is within 10 million of the ceiling.’ Do spikes count? Any ambiguity can be discussed, but would ultimately be left to the IGF or the IRT to decide.

  • Option 1 - An immediate one-time increase of the SCD debt ceiling by 20 million Dai, to a total of 120 million Dai.
  • Option 2 - An immediate one-time increase of the SCD debt ceiling by 50 million Dai, to a total of 150 million Dai.
  • Option 3 - A commitment to increase the SCD debt ceiling by 10 million each time the total supply of Dai comes within 10 million of the current debt ceiling, as detailed above.
  • Abstain - I just want to see the results.
  • Abstain - I disagree with the poll option.
  • Abstain - I don’t feel I am knowledgeable on the subject.
  • Abstain - I have a different objection.
  • Abstain - I have no opinion.

0 voters

Once again, you can vote for multiple options! Vote for all that you would vote for in an on-chain poll.

So is there an eth liquidity risk model that we utilize? If not we can can pull up the basics, like eth spread / volume trends on the “real 10” and see what numbers the market can theoretically handle under different stress tests?

Hi Patrick, welcome to the forum and thanks for commenting.

I don’t believe there is currently a risk model underpinning the current debt ceiling. To be honest I think the 50mil ceiling was a shot in the dark, and the 100mil was more of a ‘shrug lets double it.’ It was more being used to limit the total Eth at risk were the contracts to prove bugged.

Quite possibly, it would be good to get some data to support a decision.

Three days to go here, and while we have 51% on Option 1, that is not a strong consensus by any means. We also only have 27 voters, whereas we had 41 in the initial poll.

Please remain open minded about all the options. If you are not voting for Option 1 or Option 3, but you would like to see a debt ceiling increase in some form, consider voting for it to increase consensus. The golden rule is: Vote for any option that you would vote for in an on-chain poll against no change at all.


Not wanting to be that guy, but there is also the possibility of active sibyl accounts in the initial poll. Several people have voted who have no previous forum activity. While I’m not suggesting that this invalidates the poll (or even that these actually are alt-accounts) it is important to remember that this is not a secure voting system.

These things are easy to model under “normal” market conditions. I say we start by monitoring the liquidity and bid/ask liquidity ratio of the top 20 exchanges as a first simple quantitative index. The rationale is that we may need to liquidate the ceiling’s worth of eth in a short period. So it will be interesting to know the ratio of debt ceiling / total open bids. This number could be then reported weekly at the risk call. For example currently:
eth liquidity bid ~ 1.75 B
eth liquidity ask ~ 1.75 B
ask/bid ratio ~ 1
Debt ceiling to eth liquidity ratio = 100 Million / 1.75 B ~ 5.7%

How do we interpret such number (5.7p)? Is it too high? Too low? Or is just a reference point to determine if risk is increasing/decreasing?

This goes to a more general principle - let’s reduce the number of times we need to get community feedback, and let’s put an end to polls that have 99.9% support in one direction.

There are no solutions in life, only trade-offs. In this case, every additional poll presents as an attack vector, it burdens users with the participation cost. creates a potential area of dispute where none was needed (e.g. “don’t raise the SF! That’s evil, and just designed to make MKR holders rich! I don’t want to learn Econ101!”)

Let’s try to set a goal for ourselves to build systems that are not fragile against voter apathy.

3 Likes

Ideally the interpretation should be left to the individual. An index, or indicator is just a tool. Once we have a set of historical data recorded, we can meaningfully measure how we relatively compare.
I will give you a theoretical interpretation of it. At $100 million, we are at ~5% of market liquidity, and with total eth market cap north of $20 Billion, and avg daily volume of 5 to 10 Billion. we are 0.05% of the ethereum market cap, I would call it “a drop in the bucket”. If we were to liquidate all the locked eth today, you may notice a temporary spike in gas prices, and slight down tick in the price. I would conclude the debt ceiling at current levels is more of risk, because if we hit it, then SCD will rapidly go above peg and that is the real risk we should be concerned about. In a real world scenario though, things are more complicated, and it will depends on the reason for liquidation. For example if we are liquidating due to “contagion” where everyone else has to liquidate too, and it’s headline news, and cause of a bug that affects EVERY ethereum smart contract, then the theoretical analysis above would not apply. Thus, it is very easy to model under “normal” market conditions.

The latest poll has now closed. I would have liked to talk about this in the latest governance meeting, but the announcement of MCD (quite rightly) took priority.

So with that said, here’s what we’re looking at:

There is a 56% consensus to Option 3. This is the only option to receive majority support (even if only barely.) In addition, the initial general poll is sitting at 77% consensus that we should increase the debt-ceiling in some form.

With these results I think it is reasonable to move on-chain with Option 3 versus Status Quo. Here are some thoughts:

  • I would have liked to gather greater consensus that 56%, but given this is the second time the issue has come up, and given that MCD is launching in just over a month anyway, I don’t think that is a reason not to proceed.
  • The Interim Risk Team (@cyrus) has also voted for Option 3.
  • 43 and 30 voters for each of these respective polls is more than we have seen previously. This is encouraging.
  • Some voters do not have much previous activity on this forum. If you are new, I’d encourage you to take time to read in more detail, and comment or ask questions. You are welcome to vote, but if you are new there is a danger you will be mistaken for a sybil account if voting is your only form of participation.

Moving Forward

Option 3 comes with two questions to be answered, specifically:

  1. Who is responsible for setting up on-chain polls to adjust the debt-ceiling when appropriate?
  2. How long does the supply need to be within the 10 million of the debt-ceiling before action is taken?

For the sake of efficiency I would suggest that the answers to these are:

  1. The Interim Governance Facilitator (@rich.brown)
  2. Whenever the supply spikes to within 10 million of the current debt-ceiling, even if only for a single block.

I suggest continuing under these assumptions unless people object to them, so my suggestion is that this poll moves on-chain in the form below:


Title: Should we keep at least a 10million Dai buffer in the debt ceiling until the launch of MCD?

Description:

An on-chain executive vote will be created by the Interim Governance Facilitator at any time the difference between the Dai Supply and the Debt Ceiling is less than 10 million, (even if only for one block.)

The Interim Governance Facilitator should aim to create this executive vote within a week of the difference falling below 10 million.

Like any executive vote, the passing of each individual debt-ceiling increase will require the ratification of MKR Token Holders.

This mandate will expire upon the ‘official’ launch of MCD (as determined by the Maker Foundation), regardless of which date this falls upon.

Options: Yes/No


Note that @rich.brown is still at devcon and that there is a lot going on at the moment, he also may have concerns that I haven’t considered. Hopefully we’ll see action on this over the next couple of weeks.

4 Likes

yes we will hahahaha

Thanks! We are currently at 85 Million, stability fee trend is down, DAI adoption up. With engineering release dates, there is always risk of delays. I am new, so I am learning the processes. But we should have things lined up, and start voting for the increase. Let me know if there is anything I can help with.

1 Like