Weekly Narrative on MakerDAO – 06 April 2019
Weekly Narrative on MakerDAO – 06 April 2019
General:
Last week, the community continued to monitor the 400 bps increase in the Stability Fee from two weeks prior, bringing the current Stability Fee to 7.5% per annum. General concern continues to permeate as the price* of DAI still remains below its soft peg target of 1.0000 . Further, market makers as well as MakerDAO’s internal trading desk confirm that inventory levels remain materially elevated. The community has elected to continue to pursue a rate tightening policy to reduce excess liquidity in the market and remains vigilant in doing so. While the DAI price* initially indicated a reversal after the Stability Fee change and thus an upward trend, it has been noted the price is continuing to struggle and has retreated just below $0.98 . While lagging under its target, it should be noted the price* remained more or less stable.
During the course of the week, we saw an upward surge in the value of the underlying collateral thus providing more capability for borrowers to mint DAI, thus giving further rise to the concern of too much supply. Given the rapid value elevation, we saw a short-term increase in the price* of DAI due to profit taking. It should be noted that the supply of DAI remained more or less stable during value elevation of ether. Only until four days after the price elevation occurred, a supply increase of an additional (~2.5mm DAI) were brought on to the market. This minting is somewhat expected given the expanded collateral base (e.g. a correlation between a CDP DAI outstanding compared to its collateral value is not surprising). When we see the DAI supply not increasing even with an expanded collateral value, then we are quite close to correctly pricing the Stability Fee as DAI is being minted at more or less the same rate as it is being destroyed**. This is the heart of unit-elasticity.
The above being said, the price continues to lag below the target of 1.0000 ; therefore, it is reasonable to believe the supply overhang (e.g. difference between Demand and Supply) has almost ceased growing, at best it has been brought to a crawl. A removal of the *excess* Supply is needed. The main question that remains is using which approach.
In the last call, the discussion largely circulated around the following themes:
- Slow and Steady
- Bold and Aggressive
Many valid points were raised by all stakeholders as to the benefits of each path.
As a community, we must determine our core mandate and resist the notion of solving too many problems that should be left to and solved by market forces. Another way of saying the same, what is our objective as a community?
One viewpoint is that the core mandate should fall into the camp of focusing on both:
- DAI soft peg to 1.000000
- DAI stability maintaining the above with the pursuit of minimal statistical variance
As such, it was not an error to omit market maker inventory data. It is not intended as a knock on the role, but as a community we should not distinguish who is a market maker or not as at the end of the day, we are all somewhat market makers with some substantially more than others. While they have inventory on their books, the core theme should be to stabilize their inventory as we would for everyone (and not to have a protected class of DAI holders). Further, that shouldn’t imply they don’t play a critical role. They absolutely do. However, similar to the Federal Reserve which is *not* structured to save banks, the Department of Treasury does (or does not do) that, a similar ethos should apply here. The community cannot be held responsible for the profitability (or loss) of a third-party. We should only use a market maker’s DAI inventory numbers to help guide governance decisions to help modify (increase or decrease) supply thus helping to stabilize the price, for *everyone*. Market makers should start a post-mortem to determine what lead their decision making to purchase such a large supply of DAI.
Demand:
Calculating the current outstanding DAI demand is and will be the “new” scientific challenge. Further, we must start the research for the Stability Fee impact and then expand that research into its fundamental parts of DAI saving rate and the Stability Fee overall. By doing so, we will be tweak these metrics to forecast expected Demand increases and Supply decreases for when future credit bubbles occur. At present, we must resort to an educated guess and crude equations to estimate the current Demand, which must be less than the outstanding DAI of ~$90mm and more than 0. It has been estimated in the past to be closer to $70mm; however, it must be noted that with the objective to use the scientific method, this number is purely an indicative value. Several parties from the community have elected and volunteered to allocate resources to start estimating the Demand side with more accuracy.
Given the observed impacts of the Stability Fee on the price, prudence and monitoring are advised. As the overall objective is to bring supply and demand into balance (which may be observed via the price* being as close to 1.0000 as possible), it is critically important to engage pragmatically and not overshoot any new rate change adjustments.
As empirically perceived, the DAI credit bubble continues to persist after a period of low credit with lagging demand. As the Maker community is committed to a soft-peg for DAI and using market forces related to the Stability Fee to cause the enforcement of such policy, it is expected that the community will continue to tighten its policy. When the credit bubble does start the materially shrink, it is expected to see a sustained contraction of outstanding DAI (and by derivative the soft-peg should be restored), as we will have then sufficiently passed rate sensitivity elasticity of 1. When this event occurs, it will be critically important to monitor and likely soften the monetary policy accordingly based on the then market conditions with an attempt to restore demand unit-elasticity. Failure to do so can cause a severe contraction which may result in large quantity of DAI being burned rapidly.
The main differential will be how the above contraction occurs, linearly with a gentle supply contraction or an accelerated one.
It was noted in several reddit posts that 50 bps increases have had almost no impact especially on “whale” CDP holders. This point is disputed. On a percentage basis, all CDP holder feel the identical “pain” when the Stability Fee is increased. With approximately half of all of the value of DAI being minted from thirty or less CDPs, it is important to find the Stability Fee that encourage both big and small CDPs holder to being winding down their positions. If the fee is too high, we might find ourselves in the scenario where a material percent of DAI is contracted too fast as it is in the hands of a few participants.
Debt Ceiling:
As the current outstanding DAI is now around $90mm, the community should start a discussion to form rough consensus around a new debt ceiling as it is a question of when / not if the total outstanding DAI reaches the $100mm level. While not expected, a sudden increase in the underlying collateral would allow more DAI to be minted while being away from their respective liquidation price. In short, while not an urgent issue, it remains an outstanding and important issue to address. While not catastrophic if the debt ceiling is temporarily hit provided we are in an oversupplied DAI market, it would however start to artificially increase the price. Such prices increases should instead be caused by raw economic forces of supply and demand.
Recommendations:
Observe the market price* and compare the supply numbers from Maker’s internal trade desk (and any other trade desks) for another week. In parallel, prepare the community for a continual series of 50bps rate increases to continue to tighten the perceived supply demand imbalance. Further, it is recommended to continue these rate increases on a predictable and well-communicated cadence until such time as we see the excess liquidity begin and continue to contract in an orderly manner. Once we are able to effectively value the demand side, it is expected to see a trend when / where supply and demand would match. It is further recommended to keep a rolling vote & poll to continue the communication to the community related to Stability Fee changes. This has the dual benefit of both forward guidance to the community and actually implementing that guidance.
Until such time as the DAI Savings Rate is implemented and we get the DAI credit bubble under control, it is recommended to poll for the following weekly changes:
+150 bps
+125 bps
+100 bps
+75 bps
+50 bps
+25 bps
No Change
– 25 bps
– 50 bps
– 75 bps
– 100 bps
– 125 bps
– 150 bps
While the above numbers are lower than the calls for some to increase the rate multiple hundreds of basis points, the important theme is not to look for one-off solutions rather a recurring and constant rate tightening / loosening given market conditions.
Looking into the future, once we have the DAI Savings Rate in place, we will start being able to reduce the overall Stability Fee as we will no longer need to rely on purely organic DAI demand as we will be able to stimulate it via the saving rate. By directly being able to stimulate / incentivize demand AND supply with market forces we will be able to more tightly couple the overall activities into one fee. When that happens, we can bring in new voting granularity (e.g. 5 bps or lower), but we are not there yet.
Many parties have advocated for being quite aggressive in tackling the DAI credit bubble to reduce the DAI overhand. A slow and steady philosophy remains the most prudent route. Fundamentally the primary reasons are:
- Why not?
(We will get to the same elevated stability fee, just a question of when, not if)
- Who has been harmed until we get there?
(realized, not unrealized)
- Who could be harmed?
(CDP holders that historically borrowed money actually win when the price is lower than 1.0000 as they repay their loans with a discount!)
There is no dispute the price needs to go back to 1.0000 and by all accounts and given the interest on the governance call, the community is committed to get there, the question is what route, not if. If the aggressive path has systemic-risk, one must ask, what is the reward to offset that risk? There must be a reason to do so. On a purely risk-adjusted basis, there is questionable logic in material substantial stability fee changes when DAI supply growth / contraction has harmonized thus we are close to correct stability fee pricing for now** but a static supply overhang needs to be addressed. Further the slow and steady path will achieve that outcome, it just takes time to get there.
Presently in single collateral DAI, the community is hamstrung into using a hammer to swat at flies. Meaning that we can do a serious amount of damage and market changing with the intent of swatting the fly. When we should be using a fly swatter, e.g. minimal damage as the community doesn’t have the right tool, yet. The community is in need to have the DAI Saving Rate deployed urgently. Even if we have Multi-Collateral DAI but remain with only one collateral type of ETH, the DAI Saving Rate will be an essential tool. As mentioned in the recent call, the real-world corollary is like carrying an exceptionally wide tray of water (where the fulcrum is attached, for example, to a wall). The supply is controlled by lifting or lowering the non-fixed side. If lowered slowly, the water can be controlled as its volume is more or less linear. If it is lowered too much, not only does the water spill out exponentially, it also overwhelms one’s ability to hold the weight thus allowing even more water to spill out. This must be avoided, as the water in this example is the DAI outstanding. The DAI Savings Rate more or less would allow us to put the fulcrum under the middle of the tray of water with the equivalent of hydraulic jacks to raise or lower each side of the tray accordingly. In doing so, supply *and* demand may be stimulated and supported (and counterbalanced), thus minimizing the water that will be spilt.
Another primary argument against the hard and aggressive path is that the underlying market place itself is ever-changing, just as the use cases for DAI. We must avoid the scenario where we increase the stability fee to the point where the supply is contracting at an increasing pace, also known as a death spiral. In that scenario, we will have gone beyond the point where it is economically viable to create a loan. Even if the DAI price* is above 1 (where economically a borrower is borrowing at a discount receiving more in loan proceeds than they should), the issue that fewer and fewer people will want to borrow against their collateral until the rate is lower. At that stage should that occur, it is possible to tarnish **incalculable institutional reputation**.
This will be a perpetual balancing act that will become clearer (but never clear) as time goes on as our inputs will be growing (with ever present equation optimization) as compared to the traditional finance world. That said, we have a system that deals with humans, their inputs, their emotions, and their actions. No system will be perfect in automating human behavior, at least not for the foreseeable future. We should strive to have tools help guide our decision making but use that purely as a metric of consideration.
There *will* be days in the future when significantly aggressive changes will be warranted (e.g. a credit crisis). We must leave that ammunition in our arsenal and use only when needed, until then, slow and steady.
* – price being determined by USD fiat offramp via USDC – DAI (at pro.coinbase.com)
** – stability fee pricing for today’s situation and scenario, not a long-term determination
NOTE: Not a part of the Maker foundation, just my $0.02