A Framework for Real-World Assets

I believe we should create frameworks for competent managers, who believe they have the ability to generate alpha, to put their/their investors money on the line in a junior position and generate Dai at favorable rates. I think a component of this framework should be track-record, which is probably the best proxy we have for whether they’ll be a good manager.


As someone who has been the general partner in an investing LP, this is a time-honored, pretty important way of doing these things (though the devil is always in the details of how these get implemented). For Maker is likely to be especially so, since it’s hard to make someone a fiduciary on Maker’s behalf.


This is a blessing and a curse though. I am finding through conversations with potential clients, that while it’s frustrating to have to “pre-code” so much of the rules into the credit agreements, it’s a benefit to them to be able to operate with greater autonomy (i.e. within the scope of the agreement) once it’s signed.


For me number one question is inflation, given currently there’s 5% in the US and stability fee is just 3.5%… So as a base some indexation/inflation protection mechanisms and then, depending on the strategy none OR more or less aggressive alpha-generation. We could get both in e.g. equity investments of the (commercial) real estate as rents are (usually) indexed annually + the assets generate some yield.

Exactly my point! Thinking hard about a framework to own assets like real-estate equity (or ETH) seems more important than finding the best AA bond to generate 0.2% of alpha over the AA benckmark.


The good news is we don’t source our DAI from someone we must pay (leaving aside the DSR for now). And DAI are pegged to the dollar. So while large inflation lowers our profits, we are less captive to inflation compressing our net interest margin. That’s mostly because we rub our hands together and magic up the DAI, so any yield >0 can be profitable — which is, of course, not the same as paying us for the risk we assume by issuing or guaranteeing the debt in the first place.


Central banks also use magic wands to create the papers :wink: yet their no.1 objective (sometimes no.2) is price stability/inflation, so we can’t dismiss it especially while dealing with RWA.

1 Like

Yep. That’s why we have the PSM — price stability. Maker has super powers :slight_smile:

I’ve had a quick look and it appears that the primary task of the PSM module is to keep the peg at $1, so a way to keep the volatility low or non-existent, so indeed a price stability but in a sense that it basically makes sure that at any given moment 1 DAI equals 1 US Dollar.

So on the currency level it’s a great solution, yet on an investment / lending level, especially while dealing with real-world assets, it does not help much. Let’s take the current 5% US inflation as an example, and let’s assume for easier math that it continues to be at such level for the next 10 years (highly unlikely). Let’s say in the year 2020 we invest 10M DAI into X and let’s say we will get the money back in 2030 - although what we’d get back would still be 10M DAI yet what we’d be able to buy with it in 2030 is just 50% of the goods we acquired in 2020…

Of course there’s some level of stability fee and maybe initially it was thought out as a way to keep the purchasing power in place or maybe inflation wasn’t much of a problem at the DAI’s drawing table (especially while on-chain only) but in the real assets investment space inflation is quite important and has to be accounted for.

Nonetheless, even when some folks don’t pay much attention to it, there’s a certain craftsmanship that might be developed :wink: https://twitter.com/Altcoinbuzzio/status/1403870481418133505?s=20

1 Like

Not sure I get your point. If the “invested” asset is Dai, a point made by @PaperImperium is that no matter the inflation rate, any positive nominal lending rate after adjusting for risk is beneficial to Maker since there is no opportunity cost associated with minting Dai. For any other asset (say USDC), the goal will always be to choose the place with the best risk-adjusted rate. Inflation may make the eventual real return negative but I’m not sure how that factors in.


But that’s exactly what I’m after. As this is a framework discussion for RWA I’m trying to shed light on elements I find key to set up right and for the long run. Risk management is paramount but in my opinion inflation is also important - so to have a built-in preference for indexed products (TIPS vs. T-bonds; full vs. partial indexation of rents in Real Estate; etc.).

1 Like

You are right that most commercial real estate comes with built-in rent escalators for a good reason…inflation!

1 Like

Since it’s back to the drawing board, perhaps I can ask a question that I’ve been too embarrassed to ask for a while now (I’m relatively new, so I’ve missed out on so much previous discussion — sorry!):

When I first saw references to real-world assets being onboarded, I assumed they would be tokenized wrapped assets, e.g. shares of US equities, and we would onboard them like any other liquid ERC20 token. The risks we’d have to assess would be those of 1) the asset itself, plus 2) the ability of whoever is doing the wrapping (e.g. the WBTC DAO or Paxos with Pax Gold) to actually redeem the token for the asset.

We wouldn’t be inventing any new legal or technical mechanisms for this; our tried and true collateral onboarding, liquidation, etc, would work. However, as far as I can see, we don’t have any assets like this as collateral. Is that because none with with any serious volume exist for us to onboard? Has no one successfully wrapped US equities and achieved liquidity yet? (A USStocks index token seemed like a great idea, but I guess it didn’t go anywhere?)

Instead, it seems like the current RWA direction we’re pursuing is to originate loans directly to real-world entities and hold property as collateral that can’t be autonomously auctioned on a DEX nor redeemed during emergency shutdown? It seems like such a break from our current, well-oiled model that has been battle-tested for years and earned trust by holding the peg and not being hacked for all that time.

I’m sure there are good threads to read where all these discussions have happened before; can someone point me at them?

  1. I’m also not sure why it takes so much time to ERC-20 RWA (well because it’s complicated). PAXG is there, but where are stocks and bonds? So we can either wait or work already. Centrifuge might be an ERC20 way, so investing in DROP is as much a way to invest our balance sheet as partnering to build the future of DeFi. This is why the RWF CU is prioritizing more looking forward stuffs (but also exploring the trust stuff as it might be useful in any case). We don’t see investing in RWA like TradFi as a competitive advantage in the future (but something we might need to do to bridge the gap and learn).

  2. Liquid ERC20 tokens as collateral will not be super profitable. The collateral is liquid and therefore margin will be super low in the order of risk-free rate + 10 bps (as the risk is super low, liquidation was a hard problem in 2020, difficult in 2021 but will be quite simple in 2025). As a comparison, you can borrow cheaply against TSLA stock and it doesn’t even trade 60% of the time. It might work for us as we might have economies of scale. We will see. There is a reason banks don’t invest only in risk-free assets.

  3. Therefore you end up with illiquid ERC20 (or straight RWA). To us, illiquidity is defined as daily volume < 1-5% market cap (so it can still be quite liquid but not enough for a quick liquidation). That’s what finance is about, how to bridge the short-term preference of savers (DAI holders) with the long-term need of projects.

It’s not super visible, but things are moving. We have discussions with ERC20 stock index providers, with crypto-natives AAA-rated bonds issuers, … but, yeah, it’s taking just too much time. When a RWA startup takes one month to have a meeting with regulators and lawyers, a crypto-only startup can fork some code, get $2B TVL and get rekt.


Could/should we use part of our considerable war chest to fund a focused effort to make a liquid, decentralized equities exchange? Has this already been discussed and dismissed?

As to your points #2 and #3, I agree — but I sort of value stopping at barely profitable, highly liquid, completely on chain ERC-20 assets that can be managed by our battle-tested liquidation and emergency shutdown procedures. The code has been running for years and remains unhacked despite billions of dollars of incentive to do so, the peg has held, and that is (in my circles) why MKR is held in such high regard relative to any other crypto project. I think keeping that reputation backing each DAI unquestioned is paramount, while finding ways to grow DAI volume large.

I guess the main thing I’m trying to understand is, in the current RWA structure, are there still trustless vaults that people can open against the asset? My understanding is no, and that there’s going to be an entirely different, new technical (and legal) way that they function.

The idea of originating loans (especially when they’re more profitable and open up avenues for more DAI volume) sounds great, but I was hoping that maybe we ourselves could find a way to tokenize the assets, have them be redeemable for something worthwhile, and tradeable on a DEX. Perhaps the borrower would be responsible for this, maybe we’d help them, maybe through a separate entity, or something.

But then the tokenized RWA would operate within MKR’s well-established and respected risk-onboarding-vault-liquidation-ES structure.

There quite a lot of startups (or major banks) that are working on that and it’s not something we can have an edge in. I think we can help those projects as much as we can.

Correct, RWAs are more like the PSM where we define an appetite for the asset, then let people fill the vault. Possibly making a market with a spread.

I think Maker was super important to bring WBTC liquidity on DeFi. We could do the same for other kind of collaterals.

1 Like

This topic was automatically closed 91 days after the last reply. New replies are no longer allowed.