PIP 2 - Cash++ PIE

I opened a PIP for the Cash++ Pie on github.

This thread will be the place to further discuss this proposal and to gather feedback.

2 Likes

Does anybody have any strong opinions on which tokens should be in the PIE and with which weights?

I think it’s amazing to reducing risk on lending protocols. Should we consider also aggregators like Idle or it would represent an unnecessary compounded risk?

I think we should not include aggregators.

When a lending market fails often the interest rate shoots up, at which point these aggregators rebalance into them.

2 Likes

I’m also against adding aggregators.
I started working on an allocation last week, please check it out.

In short, to me the goal is to diversify the underlying risk of smart contract-based lending protocols with a weighted allocation between @compoundfinance, @Aave, Oasis (via CHAI @MakerDAO ) using the main stable coins out there:

-DAI
-USDC (@circlepay)
-USDT(@Tether_to)

Using centralized stablecoins + DAI lowers the risk associated with the peg of DAI while diversifying the interest rate exposure at the same time.

Because USDT is not supported by Compound, the weight is lower.

The exposure to protocols is nearly equal, leaning a bit more towards Oasis, according to the assessment done by defiscore.io

A Protocol gets hacked? :point_right: 66% of your position is safe.
USDT has no reserve left? :point_right: 93% is safe.
USDC starts censoring? :point_right: 64% is safe

You also get broad exposure to different interest rates without the need for arbitrage/aggregators that need to continuously rebalance from one protocol to another.

Diversification == MAGIC

A few more ideas I’m Interested exploring:

  • Allocate a 20x short to DAI to insure the peg of DAI via UpsideDai (@ChrisMaree)
  • Insure interest rate via @opyn_ (@aparnalocked)

Possible involvement of @cherry_swap @LSDefi_dao SwanDAI @savedai

Can anyone help with those? Find below the excel

2 Likes

I think using opyn for insurance would be really interesting but I’m not sure if it should be part of the Cash ++ Pie itself.

I would prefer to keep the scope limited and create the Pie with the assets @alexintosh listed in the excel and later create a PIE which contains the Cash ++ PIE and a PIE insuring the stable coins.

This way users can choose if they want coverage or not and if so even have the possibility to insure them with a different provider or a collection of providers.

I like this a lot, particularly if it’s as simple as purchasing 1 CASH(i)++ for each CASH++ you buy. Could even create a PIE of those PIEs so a single purchase can still include the insurance.

Maybe we can reach out to Opyn about doing the insurance and make them part of the conversation. Afaik they are basically put options which allow you to sell your dai or cdai for a certain amount of eth on or before a determined date.

So we would need to find some mechanism to buy the new options once the current ones expire in a way which makes sense.

Another reason why it might be a bit out of scope for now.

Opyn’s ocDAI can be minted as you like easily via
https://antoncoding.github.io/opyn-liquidator/#/

they expire in a years time from their minting date ~11 months away

1 Like

FYI, I started working on it: https://github.com/pie-dao/pie-smart-pools

Question: is CASH++ a mixture of the stable coins themselves or the tokenized lended stable coins (aka compound, aave)?

I thought about it as a tokenized yield accruing token with diversified exposure to different protocols and stablecoins.

3 Likes

Cool, so does anyone have any comments to make on initial allocation?

As it stands, this is what @alexintosh proposed with the Google Doc above:

Screenshot from 2020-03-09 21-18-50

3 Likes

Plot twist, Aave have a referral program.

Referral Program

Aave has a 20% fee-share referral program. Aave loans have a 0.25% origination fee on the loan amount. 80% of this fee is used to buy on the markets the token LEND and burn it to reduce monetary supply of the LEND token. 20% is reserved and directly redistributed to referees, in order to participate in the Referral Program, a referralCode to be used as an input of either deposit() or redeem() calls is needed.

If you wish to participate in our Referral Program, we invite you to fill our Referral program form we will provide you a referralCode in 1-2 business days.

https://developers.aave.com/#audits

2 Likes

I like the idea of keeping things simple now and trying to improve once there’s a working MVP out. Modern portfolio theory says that there is a certain amount of risk that is non-diversifiable anyway.

We could add as many stablecoin yield generating assets that exist in DeFi today, but without proper risk analysis, it’s hard to understand the value of further diversification.

I’d also like to point out that what sometimes looks like diversification might not be real diversification. For example, Compound uses Chai under the hood to juice their yields as much as possible. So in the allocation that @Mick de Graaf | PieDAO posted, ~50% of the PIE’s funds are exposed to Chai risk. Just my 2 wei on the discussion :slight_smile:

I think this looks good as an initial allocation. There is a large amount of exposure to CHAI but I think that’s ok considering the security/track record of MakerDAO. However, I think the exposure to Oasis/Chai should be amended in the excel. Very excited to see this launch :raised_hands:

image

4 Likes

I will check how we can incorporate this into the Cash++ portfolio.`

Should we redirect this fee to the PieDAO treasury or to the Cash++ holders. I think I prefer the former

1 Like

DAO Treasury makes the more sense to me.

2 Likes

@jack thanks for your contribution.

2 Likes