MIP39c2-SP16: Adding Maker Portfolio Core Unit - MPCU-001

I believe with the current setup DAI Holders can redeem for an ERC20 that represents a share of that payment stream. I agree this isn’t quite the same as we have with crypto collateral, but it’s still some sort of recourse in the event of shutdown. Previously this idea that DAI needs to be fully backed has been pretty solid among the community, so this may need some consideration (though you may just be able to adopt the mechanism used with the current RWA’s).

Given that the payment stream from US Treasury should be unaffected, I suspect we can find a way to auction the revenue stream, though I don’t have a great answer at hand right this moment.


I think what he is trying to say is that you have to be careful with the repercussions around this, as someone will probably investigate this malpractice whether it is (the sec or some federal agent) and it may affect the reputation of the protocol.

@PaperImperium can you also comment on how you expect your idea to work alongside tokenized US Treasury securities which already exist on Ethereum, especially if we were to simply onboard them as collateral and allow anyone to mint DAI with them? I also imagine ArCoin aren’t the only ones who will launch such tokens.

Assuming these are indeed low risk and the SF is lower than the returns, people would be very happy to open up such vaults.

In my example an auction of the the token in the vault would transfer ownership of the LLC holding the securities .

1 Like

Well, it’s not malpractice. These trusts exist for wealthy people to avoid taxes. It’s 100% normal, as we are not even using it to avoid taxes, just that we have no use for profits stranded outside the DAO

1 Like

We certainly can. My understanding is the yield on these is less than their expense costs.

Also, why give fees to a hedge fund when we can give it to charity?


Duplication would not be an issue as redundancy is important for a DAO anyway.

We are working on a similar idea you can check our Trust model - Work In Progress . Lately, we are aiming toward a Cayman Foundation. Plenty of questions to be answered before doing things.

Here are some bottlenecks that need to be solved:

  • interest rate risk: either you go short term (up to 2 years) and you don’t earn enough to anything significant (0.25% before all the fees) or you take interest rate and liquidity risks. If you buy 30-year term bonds you might get 2% but if rates go to 5% and you have to sell after 5 years, you are taking quite a loss or be unable to keep the peg. It’s one thing to make an experiment with People’s Company for 20M and 5-year (and/or SolarX). It’s something else to invest $1B of the balance sheet.
  • mark to market issue: if you invest in publicly traded securities you have to find a way to keep the P&L and balance sheet of MakerDAO in good shape. Not something MIP21 can do right now.

I know everyone is excited about Trusts lately, but I think the long-term play should be on-chain. Properly tokenized RWAs are coming, and I see the job of RWF to bridge the gap and work to move the ecosystem forward. There is a reason why the RWF mandate is wide for some people. Because everything is linked.

But I agree that, lately, RWA is more about reducing the PSM-USDC than the rest.


Excellent point. These would be held to maturity and do not require mark to market. Holding to maturity also insulates us from interest rate risk.

You are correct that we would need to target 7-year up to 20-year bonds at the moment, to get enough yield. The life of these trusts is capped at 20 years, so any longer-dated issuance would need to already be seasoned and not mature beyond the trust’s lifetime to avoid market and interest rate risk.

I should also note that this is a crawl-walk-run plan, and I fully expect further innovation to help us find even more flexible arrangements.

This one was chosen because it is cheap, simple, and creates a “captive” customer who distributes their profits to charity rather than shareholders.

If you’re talking about 1B+ DAI, doesn’t this constrain the DSR to be no greater than the interest rate of the portfolio? I’m not sure if this is a problem, but it’s something to keep in mind.

How are we going to decide on the charity or charities? Here are some of my favorites:

Maybe collaborate with gitcoin for quadratic voting?


Great question!

Ultimately, if the DAI is freshly minted, any positive nominal yield is good, though, as it would earn nothing at all if it didn’t exist.

Note that strictly speaking, most of the assets backing DAI already earn less than the DSR rate

1 Like

I assume we would let people vote. If we did monthly Treasury purchases, we could choose a new charity every month. I am agnostic, and would simply run a poll with people’s suggestions of qualifying charities.

FWIW, I believe you would also need a licensed broker/dealer to administer the transfer of any security in ES (likely the security would have to be distributed to whitelisted Dai holders who satisfied certain KYC and accreditation requirements).

1 Like

@omahalawyer Quick question (not requesting legal advice) – are there any tax implications of operating a Nebraska charitable trust? It looks like the trust would have excise tax obligations on the gains associated with any asset it holds (IRS statement). I am not a tax expert though.

This is a really interesting concept and, though it may take some time to get it over the hump, it’s certainly a potential pathway for this CU.

1 Like

So these will be structured as a NIMCRUT, which is tax exempt. They are required to distribute a minimum amount of their net income to a beneficiary (typically the grantor, though we will be distributing it to a charity along with the remainder at the end of the trust’s life).

Typically these are used by wealthy individuals to retain the income from an asset that they would like to allocate to a charity. Because any remainder goes to charity, it is treated as tax exempt (though it does still need to file).

This particular vehicle was chosen because for low-yield assets like Treasuries or AAA bonds, the NIMCRUT’s distributions are based on net income and not total assets. This avoids needing to sell part of the assets each year, which would expose us to significant interest rate and market risk.

Even were the trust taxed, all net income is being donated anyway. Typically only the remainder at the end of the trust’s life is donated with these, but in this case we have no easy way to get net profits from the trust to Maker. That is why Maker is simply providing a vault — to avoid KYC — and is not attempting to be the beneficiary. There are a variety of reasons why the trust won’t need to issue a 1099 to Maker – which is where it would need to be KYC’d in this relationship. The first is simply that Maker is not a US taxpayer, and 1099 is the IRS code for a US taxpayer. The second is that only most services (and the sale of fish, apparently) require 1099. The fees attached to a vault are probably able to be characterized as related to goods sold (crypto is classified as property, not currency in the US) or as SaaS, so there’s a second layer of not needing to issue a 1099, even if for some reason Maker was deemed a US taxpayer.

There may be creative ways to directly make Maker a beneficiary in the future. But the initial plan is to be very simple and giving a charity an extra $50k per year is hardly going to make us look bad.

The details sounds complex when reading on a forum, but the idea is to create our own customer. That customer can only obey rules we drafted for them, and has no shareholders who will be tempted to find ways to squeeze us once they have Maker’s financing. It’s almost the inverse of an asset originator who creates an SPV to go out and borrow – we are creating a charitable trust in order for us to approach with financing.


Tosh, a distinction needs to be made between a charity in whatever entity form that has qualified under IRC 501(c) so that donations to it by individuals are tax deductible and it itself does not need to pay any taxes, and a trust that simply has a charitable purpose and makes donations to a charity that I think could be utilized well by Maker. The former has to deal with the excise tax issue you raise (as well as sometimes having the state’s attorney general sniffing around), but not the latter. An aspect of U.S. trust taxation that can be helpful to the Maker Community in trying to diversify and invest in the real world but avoiding tax obligations is that a trust can deduct dollar for dollar from its taxable income a donation made to a charitable purpose defined in IRC 170. IRC 642(c).

That’s not to say there aren’t issues - my example of a charity forming an LLC, there would need to be certain terms in the operating agreement so that the charity doesn’t blow its 501(c) status by forming it, such as a requirement that if there was a liquidation of the LLC it would be for fair market value - the Maker online auction procedure provides that.

1 Like

I should probably note that while I am pitching Treasuries to learn and iterate, there’s likely no reason this structure can’t be used to do business with Coinbase or Genesis or a variety of other types of investments. The main limitation is having a schedule of payments known ahead of time.

But if we wanted to loan DAI to, say, Gemini where it is both insured and has a 7%+ yield or even utilize Coinbase’s offer to borrow USDC, this structure can likely get us there and do it easily and simply. The downside, of course, is leaving a little yield on the table for the trust to use.

Thanks for your proposal, @PaperImperium.

On the technical side, I’d ask you to create a Pull Request in the MIPs repo for @MIP-Editors to review.

The preamble of your proposals also needs to be changed to reflect its Request For Comments status as per MIP3.


@PaperImperium @omahalawyer Thank you both. I feel much better informed about charitable trusts in general and the thinking here in particular. Appreciate the efforts.