Liquidity Baking

Request for comments on the liquidity baking draft.


Is 2020-01-05 a typo?

Yes, fixed now. Thanks for pointing it out.

The smart contract accepts deposits of da tez and returns db tez […]

Shouldn’t that be
" The smart contract accepts deposits of da tez and returns db tzBTC […]" ?

1 Like

The contract provides subsidy = 5 tez * 60 s * 24 h * 365 d = 2,628,000 tez/a . Assuming efficient market and baking with 6% annual reward is the only alternative then we can expect liquidity provision of liquidity = 2,628,000 tez / 6% = 43,800,000 tez.

As long as liquidity <= 43,800,000 tez the yield for liquidity provision is higher than baking. When liquidity is very low, the yield is very high. Since subsidy happens every block there is no uncertainty about one’s yield. Pretty nice mechanism to ramp up liquidity.

Comparing this to Uniswap: Currently, the most liquid pair on Uniswap is WBTC-ETH with 280,000,000 USD of liquidity. With current tez price the estimated liquidity would be 43,800,000 tez * 2.5 USD/tez = 109,500,000 USD. So the subsidy alone won’t get us past Uniswap but will catapult us to the same magnitude. To make the comparison complete, Dexter recently hit “only” 1,000,000 USD of liquidity.

This is simplified/conservative: ignoring fee rewards for liquidity providers which should incentivise higher liquidity provision.


Yes, fixed. Thanks!

I think this is very useful, but could still be improved by using some decentralized version of wrapped BTC with trustless bridges rather than tzbtc.

Being able to launch this sooner might mean it’s still worth it to use tzbtc for now, but I think in that case there should be a clear plan to switch to a decentralized version at some defined point in the near future.


So I’m trying to check if I got this right:

  • Would it be correct to describe Liquidity Baker as a variation of a liquidity pool? (Or not a variation but actually a liquidity pool)
  • People will need to supply 50-50% of value in tez/tzBTC, correct? Just like in the Dexter tez/tzBTC LP. You won’t be able to only supply one of the two.
  • Everything works the same as when people currently supply value to a tez/tzBTC pool, but the only difference is the fact that they can’t delegate their tez. Instead they receive subsidy.
  • I’ve been reading through this Liquidity mining on Tezos and I’m a bit confused as to the purpose of the liquidity. Initially I thought that trading on Dexter would tap into the liquidity of the Liquidity Baker Tez/tzBTC pair. But every time someone mentions Dexter, the discussion takes a turn. Is this a correct way of phrasing things:
  1. Liquidity Baker is a fork of a part of Dexter code. So not built on Dexter.
  2. People trading tez / tzBTC on Dexter do tap into the liquidity that is provided by Liquidity Baker
  3. No other exchange or application taps into the liquidity that is provided by Liquidity Baker

Everything looks correct to me, except this part:

Dexter has nothing to do with this other than the fact that we use its contract code.

Thanks. If not for trading on Dexter or any other exchange or application, what is the liquidity for? Who uses it, what is the purpose?

Anyone can use it to trade XTZ/tzBTC. Same as how Dexter works.

Aha, so it will be totally separate. Like a single pair DEX. With it’s own interface.

Another question: the CPMM will not (yet) be implemented on protocol level. But still an amendment on protocol level will need to be made to activate subsidy, so 5 XTZ per block will be minted extra and directed to the CPMM. So Liquidity Baker could be seen as a smartcontract that has part of its functionality from protocol level?

I wouldn’t characterize it that way, but I guess? The CPMM would still function without the subsidy, there just wouldn’t be a reason for liquidity providers to use it over DEXes with higher fees.

1 Like

Sorry if this is a stupid question, but i couldn’t find an answer anywhere.

Would there be a consequence for network security?
If Liquidity Baking is too attractive, couldn’t a lot of users moving their XTZ “out of” bakeries affect Tezos overall security?

We know how attractive liquidity baking will be: since the subsidy is set to 1/16th of total rewards for a block with priority zero, it will have a higher return than baking up until 1/16th of total rolls are liquidity providers (maybe plus ~10% to account for average delegation fees). After that, being an LP on a traditional DEX that delegates performs better because the fees are higher than the liquidity baking CPMM’s nominal 1% fee.

So modulo the 1% fee and psychological effects, we expect ~6.875% of total rolls to be LPs for the liquidity baking CPMM, which is far from enough to affect network security.

It’s worth mentioning we’ve considered numerous ideas around what I’d call “decentralized monetary policy” and, aside from having the greatest near term impact, this one has by far the simplest design. The only potential security issue I see is with the tzBTC custodian, but if there are any problems 60% of bakers could shut off the subsidy with only about 1/10th of a basis point inflation.


Any chance liquidity baking gets included in F?

The TZIP has been updated. The major change is the use of the Dexter 2.0 code and a modification so that 0.1% of all trades are burned, which at 7.2mm daily tez volume offsets inflation from the subsidy.