Skip to content

Latest commit

 

History

History
106 lines (85 loc) · 8.75 KB

risk_17.md

File metadata and controls

106 lines (85 loc) · 8.75 KB
Raw

Risk #17 - 1/3/2019

Introductions

  • Rich, Steven

Recap

What do risk teams look like? What should they do? [00:48]

  • There’s a spectrum of things risk teams can contribute to the governance process
  • In the end both teams should submit

What is a supporting team? What do they do? [1:37]

  • Non-risk parameter, things like: Bring in data, clean data, model validation service, quant or qual data, recourse audits, technical audits
  • Cont’d below

What does a recourse audit entail? (DGX token holder fly to SG to recover gold bars) [2:44]

  • If everything fails, how do we go back to the original issuer and make a claim against them?
  • Gold bullion may not be the end recourse, would need local currency like Singaporean Dollars
  • Liquidating bullion into fiat is well documented and regulated through brokers, not as simple
  • Audit describes the path to recourse. A similar process can be used for NFTs, invoices, etc

What collateral-specific recourse are we auditing for? [6:38]

What is a supporting team? [7:01]

  • Supporting teams have to cover a wide spectrum of risk verticals
  • We need to consider:
    • How do we vet these teams?
    • How do they contribute?
    • How does their contribution get compensated?
  • Supporting teams may participate in the submission process by submitting their work that is being passed to the other risk teams
  • Ex. A risk team cleans data and provides it to the system.
    • Should we require the data be open sourced?
    • Do we allow them to keep the data and share with the risk teams that need the data?

Where do risk teams come from? [9:27]

  • We’re adopting traditional tools and applying it to the space with appropriate tweaks
  • The decentralized nature of the function being the largest tweak

Are there agencies that can be contracted to do this work? [11:25]

  • Absolutely. Big 4 auditors have departments to independent companies and investment banks, etc
  • The detailed processor everything have been documented
  • The blockchain gives us an advantage on the transparency side

What do we do with respect to the data these teams create? How do we compensate them appropriately? [13:36]

  • When a team takes on the process of cleaning data and making it available the MKR holders, through the governance process, need to vet the process
    • Where are you getting the data?
    • How are you cleaning it?
    • How are you providing it?
  • Is the data kept within the risk team or made available outside?
  • With respect to both qualitative and quantitative data
  • Do qualitative teams submit to the more quantitative teams wrt final submission

What mechanisms are in place to support these processes? [18:07]

  • We’re diving into the inner workings of the risk process so that the community might find opportunities to grow the ecosystem through new businesses
  • The internal risk team will be the model for the path to becoming a risk team, being vetted by the community, being accepted into the system, and submitting parameter proposals for voting

In addition to collateral models, we’ll need two other types of models [21:16]

  • Portfolio models - looking at the makeup of the portfolio as a whole
  • Economic models - updating parameters like the stability fee

Risk function will come front and center, more than the previous year [24:27]

  • More questions will arise like:
    • "How do I become a risk team?"
    • “How does one submit risk parameters?"
    • “What does the data look like?"
  • Potentially answering those questions as well.
  • The internal team is running parallel to reaching out to the community, to get involved in the risk process, will be releasing a Liquidity-adjusted Value At Risk assessment
  • Rather than providing a list of the process steps and inevitably attempt to discover all the nuances, working with the community through the process the nuances can be discovered organically

What are some of the things risk teams are looking at if they’re taking a holistic view when a new collateral type is proposed? [33:29]

  • In the traditional finance world there are expected risk, expected returns, correlation matrices
  • The liquidity adjustment comes into play wrt allocating large positions slowly to prevent slippage
  • We have the three debt ceilings
  • Consider the collateral’s business, economic function and the remedy to any dislocation of supply and demand

Digix for example [36:41]

  • Bringing in Digix even though the market cap is only 3 million may catalyze the demand for DGX

Does that potential inflation of market cap pose as a risk to the system? [38:35]

  • We’re driven through two primary functions: the ability to pledge collateral and the ability to draw value
  • Want to avoid being the primary factor of inflation
  • The Maker system will drive significant value to Ethereum and other networks, given interoperability

Could this contribute to a catastrophe if the value is derived from being a collateral? [41:07]

  • If mismanaged, absolutely
  • Banks will lend to small cap companies at a 10:1 buffer to be safe
  • The bank will look into recourse. Determine the value of the assets or take the stock and find a buyer of last resort

So, when a collateral is added the liquidity risk will be factored in up front? [43:07]

  • Yes, the analysis will need to be ongoing to ensure continual compliance
  • That is our competitive edge

In the traditional market when you’re trading futures there is a basis risk that describes the difference between the future price and the underlying asset price [45:12]

  • There is similar application to tokens. What is the basis risk between the token and the underlying company?
  • Security token: either a token to represent a company share or a token to represent another security
  • Hopefully these ERC assets retain the correlation and volatility of their underlying value rather than succumbing to the crypto volatility

Would we accept short tokens like those from DY/DX as collateral? [48:18]

  • The acceptance of a collateral boils down to the ability to recourse all the way back to the liquidation preference
  • The primary piece of recourse we have right now is price discovery in the open market