Reverse Continuous Auction on Bonds


Hi everyone,

I have been going around the model of CO and had a couple of questions before. Would love to hear your thoughts on this @thibauld.

In this post, I would like to share a simple investment process based on bonds (loans) instead of tokens which, I feel, has some similarities with the CO concept, although it would be different in some aspects.

The process would be as follows:

  • Anyone can offer to loan money (DAI) to the project by transferring it to a smart contract and specifying a risk bonus required in exchange. For example, an investor can send 100 DAI and sets his risk factor to 20X. Many investors can make different loan offers.

  • The project will then consume these offers as it needs cash to operate. It will naturally select the cheapest offers first (with the lowest risk factor). When the project consumes an offer, the loan obligation is effectively acquired to return the loaned amount times the risk factor. For example, if a 100 DAI at 20X factor is used, the project commits to pay 2000 DAI back to that investor. Not consumed offers cab be withdrawn by investors at any time.

  • When the project generates revenue, this revenue is channeled through the smart contract which will automatically start paying the loans back.

This is a relatively easy to implement smart contract, and all the off-chain trust would be focused on the fact that an investor won’t be sure the project will channel revenue to the contract in the future. However, it is possible that strong pressure will exist on the project to honor that commitment since the public address of the contract might become widely known (could even be the project address in ENS), and so that asking the consumer to pay for services to another address would clearly signal fraud by project owners. Standard legal contracts signed by the project owners at the moment of investment receipt could also be used to enforce this.

I like the simplicity of the process and its flexibility to adapt to different uncertainty/risk conditions. I also like the fact that, at some point, all debts are paid back, and the project may become public property.

I think flexibility in risk appreciation is important because, even if it might seem natural that risk factors should go down as project evolve and deliver, this is not necessarily the case. Bonding curves seem to hard-code this uncertainty reduction in the fact that new investments receive few FAIRs.

I also like that it has a similar already existing legal instrument which are convertible bonds, which could perhaps simplify its legal compliance, but hey, the securities ghost will appear there.

So any thoughts? Do you see big differences between this approach and CO? which ones would these be?

You can read more about this here