Can the DAO "restrict" itself?

Let’s say MKR holders wanted to promise Dai users that they won’t hike the DSR spread for a certain amount of time.

So they said “We will keep DSR spread at 0.5% for 6 months”

What are the various ways they could go about to promise this? Meaning how could they restrict themselves from changing the spread?

Even if hypothetically a 6 month cadence was set in place, what’s to stop someone from submitting a signal request to reduce the cadence? Then the promise would be rendered useless

Option 1: Smart contract that does not allow votes on DSR spread to be changed.

Problem: Someone who works for the maker foundation told me this impossible from a technical perspective

Option 2: On-chain governance poll that says “Set DSR spread, and do not allow it to change for 6 months” then no one will submit a signal request to override it for that time

Problem: What if someone creates a brand new governance dashboard and disregards the old poll and starts submitting malicious executives and somehow gets MKR holders to vote on it

Option 3: MKR holders put their assets in a contract that says if they change said parameter their holdings will be taxed

Problem: I don’t think anyone would actually do this

What do you all think?


Option 1 might be possible with a DSR-specific GSM, but I think governance could always vote to remove that delay (but that action will have its own delay. So the counterparty will have time to find out that Maker is reneging on its word).

Option 2 is most likely. Just social enforcement. An alternative dashboard that is not trolling or malicious will likely want to honor the credibility of Maker, so I don’t think that’s too worrisome.


No, we can’t really do this currently, which is a shame, because I agree it would be useful.

Option 1 doesn’t seem promising.
Option 2 seems workable, but not ideal.
Option 3 seems doable, but requires dev work.

How about Option 4? Maker Protocol (via executive spell) creates a prediction market on Augur v2 and places a 100,000 Dai bet against the DSR changing in the next six months. In practice, this is staking Dai against it changing. This gives us some very cool things:

  1. Gives us a credible pre-commitment backed by Dai.
  2. Allows us to see how trustworthy the DAO is perceived to be by the market as a whole.
  3. Gives us another source of income from the trades on that market, and potentially winning bets.

I think this “restricted” case will be the default when the GSM starts getting extended to weeks/months and the Instant Access Modules come online. Essentially hardening the protocol against any sudden unexpected changes in general.


I wonder if this could be modeled as some sort of future contract? Like lock in your 100 DAI today and in return you get a contract that says in 6 months you are entitled to say 103 DAI. Once the contract is issued the 100 DAI would get locked in the DSR.

Now assuming 8% APR you would assume that the DAI would be worth 104.07 DAI in 6 mo time, so the MKR holders stand to gain 1.07 DAI from this arrangement (if they keep the DSR at or above 8%). If the DSR drops below 8% though they might end up on the hook for the foregone interest which would presumably result in the difference being added to the debt buffer.

In the meantime though, the person offering the 100 DAI gets the peace of mind that he won’t need to worry about fluctuations in the DSR and can just count on his extra 3 DAI being there in 6 months time.

Disclaimer: I have no idea how technically feasible this would be, just an idea for how you might go about this.

1 Like

Hi @Akiva

I think it would be possible to nearly guarantee fixed interest rates using these guys

How it works: your locked DAI has a variable revenue stream based on the Dai Savings Rate. You use Tinlake and split this revenue stream into a fixed part (Tin) and a variable part (Drop).

You wanted stability so you keep Tin and sell Drop in order to compensate for the lower returns of the now fixed stability fee.

@spin how doable is this with some minor adjustments of Tinlake?

1 Like

@LongForWisdom I was talking about the DSR spread not the DSR itself.
I don’t think Augur is able to solve this. If there is a vast number of people wanting to hedge spread, who will be the vast group insuring it?
Of course there could be insurance companies but I’m not sure that will happen anytime soon.

BTW re hedging DSR, (not spread) my company has a fixed DSR solution that we’ll be launching very soon. Currently its in the third independant smart contract audit.

Yes upon reflection I agree that social enforcement would be the most likely, but it introduces new “trust”

1 Like

Interesting. It seems i too have misinterpreted the original question. Would you care to expand upon your interest in the DSR spread as opposed to the DSR itself?

1 Like

The DSR is possible to hedge against. All you need: for every Vault holder with 1 Dai drawn - is 1 person who wants DAI with a fixed rate. If you combine Dai with a vault and balance them out you end up with stable rates on both sides.

Here are some pictures to explain

As you can see in the second image, the size of the difference between forward rate and DSR is the size of injection to or extraction from vault.

However this approach leaves the DSR spread completely unhedged. I’ve spoken to some experts about this and they believe the best way it to have a poll and then for MKR holders to have integrity that they won’t change it.

I can understand the desire to want a flexible spread during times of uncertain DSR utilization, but once that stabilizes I think a precedent of low volatility spreads should be in place.

1 Like