Weekly (almost) Narrative on MakerDAO – 10 June 2019

General:

During the course of the last week, the community voted on and reduced the Stability Fee to 16.5% per annum. The community has maintained the polling frequency for another executive vote. At present and pending an executive vote, the Stability Fee looks to be maintained at 16.5%. We have seen some unique voting trends whereby voting apathy or fatigue has appeared when a vote is exceptionally unlikely to implemented in an executive vote.

The overall price* of DAI has solidified around its soft peg target of 1.0000 and for the most part stuck to the peg. During the last week, the total outstanding DAI has now expanded to just slightly above 82.3mm DAI.

The above being said, as the DAI price* continues to hover right at the target of 1.0000, we can draw an initial conclusion that most of the market maker inventory has been cleared out; therefore, it is reasonable to believe the supply overhang (e.g. difference between Demand and Supply) has ceased growing and is basically zero. Further monitoring is advised as without sufficient market maker inventory, the system has very little ability to absorb purchase shocks should a market participant decide to purchase DAI in volume.

In parallel to the above, secondary lending markets continue an identified trend to lower their rates somewhat on par with Maker. As discussed in the prior narrative, the spread between the average of the secondary lending markets and Maker will continue to be compressed as Supply and Demand are brought more into harmony. Important note: the aspect with secondary lending platforms will morph as new collateral types are on-boarded. The expected rate compression should be only compared with a similar collateral type, in this case, ETH.

Further, the average daily maker burned (as calculated) is now right at 50 MKR per day, down from over 60 after the recent Stability Fee decrease. The total MKR in the “burner wallet” has now surpassed 1609 MKR. As seen from the previous week, only 50 MKR have been burned in a timeframe that should have seen 7x the amount. The P/E ratio (fully diluted less the burner wallet) has also increased as a result of both price appreciation of MKR along with the earning component bring reduced with the recent decrease in the Stability Fee.

Demand:

With the upcoming introduction of the DAI Savings Rate, we need to start polling for / forecasting where to start the DAI Savings Rate. As it is strongly recommended to treat the introduction no different than another other new market force, it is strongly advised to roll-out the DAI Savings Rate slowing starting at 100bps. As the DAI Savings Rate should be viewed as a competitor to traditional saving rates or even United States Treasuries, iterations on the increase post DSR launch should be no more than 25 bps at a time in general (but as further outlined and determined / calculated below which may be less than 25bps).

Referenced Presentation Link (apologies in advance for the manual drawings):
https://docs.google.com/presentation/d/1t0uuDfWwUvCwN2Ot5pchxX1TmCrVv54QVzsPdwWX58I/edit?usp=sharing

In the previous governance call, the topic of risk was discussed at length and more specifically how to approach each type of risk as we start the discussions of adding new collateral types. Systemic / Depth / Probability / Ability to liquidate / Slippage / Liquidity of the market, etc.

As a function of general risk management, upon Multi-Collateral Dai launch, it will become necessary to evaluate all aspects of risk for a given collateral including its interactions with other collateral types.

As the Stability Fee for collateral #1 (“SF1”) shall be computed with the following equation:

SF1 = DSR(uniform) + Oracle Fees(1) + Risk Team(1) + VaR_MKR(1)

In this scenario, VaR_MKR(x) is being based on what does it takes to “insure” 95% of the value against loss based on that collateral’s statistical probability of loss after liquidation, basically insurance premiums (as VaR_MKR with 100% would require the “premium” to be the same value as the asset itself as correlation between 0 and 100% is not linear).

(Note: While not discounting their value to the system as a whole, for the purposes of this evaluation, we are going to negate the Oracle and Risk Team fees as they should be materially close to zero and thereby negligible when viewed from a macro perspective.)

As an MKR token not only represents a governance token that votes, it also represents a portion of a quasi-insurance fund. For each good loan that is repaid with interest (where the VaR_MKR portion buys and burns MKR) the reverse holds true that for each loan at a loss, MKR is diluted accordingly to cover any losses. For MKR to set a VaR_MKR(x) price that has zero risk to MKR holders, it would have to require the same value as the collateral itself. Like a $1mm life insurance policy that is priced for a healthy person at age 18 (which will be pennies on the dollar) the closer a person gets to their statistical death target, the higher that certainty becomes (e.g. pricing a $1mm life insurance policy for a 100 year old is going to be really close to $1mm).

At present and without MCD, the SF1 purely equally the VaR_MKR(ethereum). While obvious and redundant, we know directly that ethereum was used as collateral to mint DAI. Unless we intentionally analyze and identify which collateral is being used to mint DAI in an MCD world, the signal will be lost and obfuscated by the noise of DAI being minted based on a variety of collateral types.

Logical progress:

When MCD launches, it is recommended to do so with either no other collateral types other than ETH and the DSR …. OR…. with two collateral types and NO DSR for a while. It is strongly recommended to establish a baseline for analysis purposes. Thereafter the DSR can be introduced in a pragmatic way.

Let’s take an example where we have two collateral types and the DSR, (purely hypothetical), in a world where we have two collateral types, ETH & REP Imagine a situation where we establish a $200mm debt ceiling for each collateral type, the DAI Savings Rate is set to 100 bps, and we further set the underlying VaR_MKR(1) and VaR_MKR(2) at 16.5% and 20.0% respectively. Imagine we do so, and DAI is minted at a fierce rate with 1mm being minted from ETH as the collateral and 3mm being minted from REP as collateral thus causing the DAI price as a whole to be degraded to $0.97. How should reasonable governance respond? What are the prudent steps?

The answer isn’t so clear or straight forward as we must first decide for each collateral type where each sits as a ratio of DSR / VaR(x). Meaning that when we attribute certain activities to a collateral type that we must allocate a weighted model to how it both impacts the collateral and how it impacts the system as a whole, as the DSR impacts all asset classes, not only one. For this scenario, as we start the underlying VaR(1) lower than VaR(2), it implies that it is a less risky asset.

So back to our scenario, one could conclude that since the price is below $1, monetary policy action is required. Further, one option would be to simply only raise the DAI Savings Rate (which crosses all collateral types) to cause the absorption of the excess supply thus bringing the Supply and Demand back into harmony and thereby restoring the peg. However, by doing so we would not be properly allocating the risk for a “risky” collateral (or more specifically the risker collateral of the two). In the scenario as outlined above prior to be added as collateral, we / a risk team needs to identify as a base case what is the default allocation of monetary policy change between the DSR and the VaR_MKR(x) per collateral type. This ratio for the most part should be static with maybe annual or bi-annual updates. This means that for ETH like REP, we would probably allocate 5% to the DSR with 95% to the VaR_MKR(x). So in the above scenario, it would make sense to proportionally raise the DSR from 100bps to 105bps while changing the VaR_MKR(1) from 16.5% to 19.4% and 20.0% to 24.7% respectively (not actual numbers but just conceptual; however, the point is that the more risky asset that minted disproportionately more DAI should have its rate elevated more than the less risky asset that minted less).

The point here is that the DSR impacts everyone. This introduces a quasi-game theory Nash-equilibrium as each change to any VaR_MKR(x) (which also has a corresponding DSR change) impacts everyone. On a weighted calculation some collateral types should experience lower costs and some greater. Similarly, the DSR will be pulled up by some and pulled down by others. So while the DSR initially was viewed as somewhat the savior to elevated interest rates, it cannot remove the risky nature of the underlying collateral; therefore the rates associated with that collateral should be elevated until such time as the collateral is no longer as risky.

Now let’s reference an asset that is as close to “risk-less” as possible, a US Treasury (or simplicity sake, lets just use a striped version, so no coupon). In the inverse manner with a UST being riskless (almost), when used as a collateral base and when large quantities (for example, 5mm) of DAI are minted because of it, changing the VaR_MKR(x) for a UST would not make sense. Rather the weighting between the DSR and the VaR_MKR(x) would look more like 99.9% DSR and 0.01% VaR_MKR(x) .

As we introduce a new collateral type that is more risky than a Treasury and less risky than ETH, we need to start a discussion on what percentages to allocate to the DSR and which to its VaR_MKR(x). For example, let’s evaluate the public debt of a “BB” rated company that has been tokenized in the form of an ERC-20. Here, the collateral is not “riskless” but something that has far less variability and less risk than ETH. As such, an approximate allocation of 95% DSR and 5% VaR_MKR(x) probably makes sense.

So now if we integrate all of the scenarios together, we have ETH, REP, UST, and BB rated debt (each minted DAI 1mm, 3mm, 5mm and 5mm respectively as purely an example). When we have large DAI minted on aggregate, we have no clue what to do without analyzing the data to where the DAI was sourced from (or burned). Thereafter, when we know what the DSR / VaR_MKR(x) ratio is, we will be able to have some vision in what the logical steps actions would be. As such and without doing the simultaneous equation math, the summary is that the DSR should be disproportionately used for the UST and BB minted DAI while elevating the ETH and REP VaR_MKR (with REP being elevated more than ETH).

The process to determine the exact starting percentages should be purely a scientific one based on probability of the collateral materially degrading in value when viewed in isolation. Thereafter, that percentage should not be set in stone but in a similar light to a collateralization ratio before a liquidation, they really should remain unchanged until such time as the data supporting science demonstrates that the VaR_MKR(x) for that asset has also materially changed (improved / degraded). This is no different than S&P rating a company / country with a higher credit score and thereby they have access to lower costs of capital (while differentiated short-term and long-term). The same should hold true as well for Maker and its risk-teams. As such, it is recommended to introduce this concept when evaluating each asset that is considered and then implemented as collateral.

A governance question is now raised (and a potential attack vector) on who sets those ratios? Who updates them? How frequently? Does the community vote those changes in? Or does the community vote in those that make those recommendations? A valid argument can be made for each one with positives and negatives for each.

This becomes an essential multiple simultaneous equation multiple variable parallel math project. Further, this challenge is then amplified with the objective of using a PID to help algorithmically make recommendations for the community as the pace at which capital (DAI) is deployed might very well be lineral or could be exponential for a while and still be healthy.

In contrast to some prior narratives, in an MCD world with only ETH as collateral the DSR will slightly increase. However, upon implementing the second (and subsequent) collaterals, it becomes important to only change the DSR in a weighted model (e.g. ETH shouldn’t get special treatment).

This will require restraint and discipline from the community, as it will be quite tempting (with a risky asset) to simply raise the DSR (to kick the can down the road) to absorb excess DAI while artificially leaving the VaR_MKR(x) artificially low. By not correctly pricing the VaR_MKR(x), we would be introducing excess (and deferred) risk to the system. Furthermore, by doing so, we would have the side-effect of pricing the system out of low-risk assets (as their corresponding DSR +VaR_MKR(x) would be disproportionately (and unjustly) elevated).

The act of kicking the can down the road would be directly felt in the future when lower risk assets would be added. How could we reasonably price the risk of JPM debt close to ETH? Where would it fit in?

Recommendations:

It is expected that the DAI Savings Rate contract will be an essential metric to gauge and estimate demand changes. As such, it is important for future PID algorithm optimization to gather as much demand side information as possible, hence the logic to start slow and iteratively determine the market correlation and sensitivity.

As a community, we need to learn / decide / re-apply what we consider a healthy market? Large volumes of Risky Collateral / Riskless / Hybrid? Supply & Demand always in harmony? Age of Debt? When deployed, is a collateral healthy if its DAI growth is linear? (or replace linear with expected?) How do we model and incorporate “expected”?

As each time we introduce a new asset type / class / “species”, the DSR and the VaR_MKR(x) will need to update the parallel calculation and the PID as well. In a continuation of the general theme, it is recommended to establish a workgroup to start the logic on a Machine Learning PID Control System** tool to help with signaling. A small group is already coming together with help presently anchored by Vishesh, Alix, and me. That said, the challenge is ever present and impacts all of us. Please connect with us to volunteer some cycles on Rocket Chat or on Telegram.

As an aside, many of the points raised above are in direct contrast to what was promoted not even a month ago in prior narrative(s). Learning requires one to make mistakes and adapt. No doubt my list of mistakes (aka learning opportunities) will continue. In a world that seemingly criticizes a lack of perfection, yeah… definitely not perfect here… Just learning… Rinse and repeat…

• – price being determined by USD fiat offramp via USDC – DAI (at pro.coinbase.com)
  ** – PID (proportional / integral / derivative) Control System. It is expected that the DAI Savings Rate will be a potent (and more so than the Stability Fee impact on Supply) tool, as such it will be supply leaning (much like an autonomous car that has an alignment issue and drifts to the right, the control system for Maker will unlikely be symmetric between supply and demand).

NOTE: Not a part of the Maker foundation, just my $0.02 and not intended as advice in any capacity.

Top 250 MKR holders = 688800.833
1d 🔺: 257.860
1wk 🔺: 452.809
Live STBLTY Fee: P/E (dilut.) 53.02 – P/E (w/o dev. fund) 39.27
FCST 50bps STBLTY Fee (VaR MKR burn portion): P/E (dilut.) 1758.93 – P/E (w/o dev. fund) 1311.20