Next-generation stablecoin design: How to balance the impossible triangle?

Original Title: "Design of Next-Generation Stablecoins"
Author: DODO Research Institute

The Impossible Triangle Argument

The impossible triangle constrains the credit currency system of sovereign nations and has been repeatedly proven in practice. Stablecoins also face the impossible triangle and are subject to its constraints. Therefore, before discussing the design of next-generation stablecoins, we need to understand the impossible triangle and establish a connection with stablecoins to grasp the key to stablecoin design.

What is the Impossible Triangle

The impossible triangle was independently proposed by Mundell and Fleming between 1960 and 1963 and has been repeatedly proven in the practice of international finance. The impossible triangle refers to the fact that a sovereign nation cannot simultaneously achieve a fixed exchange rate, free capital movement, and an independent monetary policy.

• A fixed exchange rate system means that the exchange rate of the domestic currency remains stable against a certain currency, thereby achieving stability in the value of the domestic currency. Fixed exchange rates are beneficial for the stability of import and export trade.
• Free capital movement means that there are no restrictions on capital, allowing capital to enter and exit the domestic financial market at any time.
• An independent monetary policy means that the central bank of the country can freely decide what kind of monetary policy to implement, control the issuance of domestic currency, and determine the domestic interest rate level.

Why can't the three elements be achieved simultaneously? Simply put, for a sovereign nation's credit currency, if capital is freely movable, an implicit condition is that when the yield provided by the sovereign nation's credit currency system is no longer attractive, unrestricted capital will completely withdraw. If, under free capital movement, a country's central bank implements an independent monetary policy, considering scenarios such as increasing currency issuance and lowering interest rates, capital will withdraw from the country's financial market due to falling interest rates, leading to significant selling pressure on the currency and depreciation pressure. To maintain a fixed exchange rate system, a large amount of reserve assets must be available to absorb the selling pressure, which is often unsustainable compared to international speculative capital.

The Impossible Triangle of Stablecoins

Many researchers and scholars have their own interpretations of the impossible triangle of stablecoins, but there is no consensus on the expression. This article will combine previous research and discuss the impossible triangle of stablecoins based on its own judgment.

This article believes that the impossible triangle of stablecoins includes stability, decentralization, and capital efficiency.

Decentralization

What is Decentralization

In the context of stablecoins, decentralization has two meanings: transaction decentralization and asset decentralization.

Transaction decentralization refers to transactions being unrestricted, where no entity can prevent transactions within the system. The lower the degree of transaction restrictions, the higher the degree of transaction decentralization. Asset decentralization means that reserve assets cannot be controlled or misappropriated by specific entities; the lower the degree of control by specific entities over reserve assets, the higher the degree of asset decentralization.

The degree of transaction decentralization has already been quite high at this stage. Many stablecoin projects rely on public chains to achieve transaction decentralization, and the highly developed exchanges make capital flow nearly completely free. In fact, within the realm of digital currencies, the free entry and exit of global capital has reached unprecedented heights.

However, asset decentralization is difficult to achieve. The reserve assets of USDT and USDC are entirely controlled by the project parties. Even with transparent asset disclosures and third-party audits, the risk of asset misappropriation by project parties cannot be eliminated. Even projects that control liquidity through protocols have examples of project parties misappropriating reserve assets. When Luna collapsed, LFG had reserved a large amount of high-value reserve assets like BTC and AVAX for UST; whether these assets were used to stabilize the market remains opaque. Solutions for asset decentralization are still a challenge.

Capital Efficiency

What is Capital Efficiency

Capital efficiency encompasses many factors, including the growth rate of the system itself, lower transaction costs, and the yields of certain protocols related to the system. However, these can all be discussed as independent monetary policies. Why? Because capital efficiency, as a constraint of the impossible triangle, actually refers to whether the system itself can independently regulate capital efficiency. The collapse of Luna illustrates that independent high yields cannot be maintained while simultaneously achieving stability and decentralization. Once expectations change, capital can freely withdraw, leading to severe selling pressure. Therefore, if stability and decentralization are to be achieved, capital efficiency becomes unsustainable.

Over-Collateralization is Not a Loss of Capital Efficiency

A common misconception is that over-collateralization is a loss of capital efficiency, and that efforts to improve capital efficiency should focus on lowering collateral ratios. This is a misunderstanding. Whether or not there is over-collateralization is unrelated to capital efficiency; over-collateralization merely reflects unreasonable credit expansion and indicates a reasonable exchange value of a new form of credit at a specific stage, rather than a loss of capital efficiency.

From a long-term perspective, during periods of rampant credit expansion, people will find over-collateralization rates glaringly high, leading to a pursuit of higher leverage. However, when the cycle reverses, a collapse will inevitably follow. It is challenging to separate the process of strengthening credit from capital efficiency because when a new form of credit is rapidly embraced and expands outside the system, the returns from external capital entering will exceed the internal rate of return. At this point, it becomes difficult to discern whether high returns are driven by capital inflow or credit expansion, or by internal returns.

The only way to address this is to adopt over-collateralization for new forms of credit to avoid uncontrolled credit expansion. Rapid credit expansion is detrimental in the long run, although it may lead to quick wealth redistribution in the short term. Over-collateralization effectively reduces credit risk and is a necessary choice for new forms of credit.

Stability

What is Stability

Stability refers to maintaining a fixed exchange ratio between digital currencies and sovereign currencies. Stability can be divided into long-term and short-term. Short-term stability can naturally be achieved through reserve assets to offset selling pressure, but reserve assets will eventually be depleted. To achieve long-term stability, one must inevitably make trade-offs among the other two elements. At this stage, stability relies on sufficient reserve assets.

Credit forms are not static. The previous attempts at algorithmic stablecoins aimed to create a new form of credit, but it is impossible to create a new form of credit out of thin air; new credit forms need to rely on old credit forms and gradually transition.

Design of Next-Generation Stablecoins

The issues to consider in the design of next-generation stablecoins are how to make choices among the impossible triangle, how to marginally expand credit. This article proposes a retail model for credit expansion to address the marginal issues of credit expansion, introduces a dynamic balance of the impossible triangle to make choices based on circumstances, and proposes a monetary policy regulation framework to control currency issuance.

Some research institutions and scholars have also proposed alternative design ideas, but they focus on specific issues such as reserve asset selection, collateral ratio selection, and stability mechanism construction. There is relatively little discussion on how stablecoins can expand credit and how to choose among the impossible triangle, which is crucial for the long-term success of a stablecoin project. Therefore, this article proposes three areas for improvement in the design of next-generation stablecoins from these perspectives.

The stablecoins discussed in this article are digital currencies aimed at stability and do not specifically refer to stablecoins backed by certain assets or algorithmic stability.

This article believes that the design of next-generation stablecoins should pay attention to three points: in the choice of the impossible triangle, achieve dynamic balance among the impossible triangle, adjust choices based on market conditions, and respond flexibly; adopt a retail model for credit expansion; and in monetary policy, the native token should partially absorb seigniorage and build a liquidity market.

Dynamic Balance of the Impossible Triangle

For a stablecoin project, the constraints of the impossible triangle are not rigid; that is, the three factors are not mutually exclusive rigid constraints. Each factor can be chosen with different intensities. A more reasonable expression of the impossible triangle is that the three cannot simultaneously achieve high intensity. Stablecoins can choose different degrees for the three elements at the same point in time and can also adjust the strength of the three based on different stages of development.

Same Point in Time, Different Degrees, Non-Rigid Elements

Regarding stability, it is important to note that stability itself is not sufficient to encompass all choices; stability itself also has varying degrees. By selecting collateral ratios, reserve assets, and controlling the friction costs of redemption, a stablecoin project can exhibit different volatility and acceptance capabilities. Choosing stronger stability will inevitably weaken the other two factors.

Similarly, decentralization works the same way. The higher the degree of asset decentralization, the poorer the flexibility in adjusting assets. The higher the degree of transaction decentralization, the freer the capital flow, making it more difficult to control the outflow of assets from internal to external and the inflow of assets from external to internal, thereby affecting the other two factors.

The stronger the independence of capital efficiency, meaning the more independent the monetary policy, the more it requires regulating the internal yield of the system, which demands a more closed system under stable conditions and a more fluctuating exchange rate under an open system.

In summary, the three elements are not rigid; one can adjust the relative strength of the three to achieve balance.

However, realizing that the three elements are not rigid actually makes the choices more flexible. This is reflected in both RMB and FRAX. The RMB allows free movement of CA accounts (current accounts for trade in goods and services) while strictly controlling KA accounts (capital financial accounts). Although the RMB is a fixed exchange rate system, it also has a floating mechanism that adjusts the target price based on circumstances, and the independent monetary policy is not entirely independent, as the Federal Reserve's interest rate hike and cut cycles are also considered.

Dynamic Rules

Although the range of choices has become broader and more flexible, how to adjust based on external conditions is a more critical issue. This article provides dynamic rules to guide the dynamic balance of the impossible triangle at a macro level.

For the choice of dynamic balance, the basic principle is to comply with the large cycles of traditional finance and the development strategy of the system itself, but to be more aggressive during expansion periods, as the entire digital currency market will also have significant external absorption. The macro choices include strengthening internal absorption, optimizing internal growth, and speculative products.

• Strengthening external absorption: high interest rates (capital efficiency), anchoring (stability), restricting free capital movement (locking, etc.)
• Optimizing internal growth: low interest rates (capital efficiency), anchoring (stability), allowing free capital movement
• Qualified speculative products: high interest rates (capital efficiency), floating (stability), allowing free capital movement

Using a Credit Retail Model for Credit Expansion

Credit expansion for a stablecoin project is independent of the impossible triangle choice issue and is even more important than the impossible triangle choice. Credit expansion is a model system for how stablecoins can issue currency and improve their credit level at a micro level. This article proposes using a credit retail model for credit expansion, which has flexibility at the margin.

In the traditional sovereign credit currency system, credit expansion relies on a complete set of central bank and commercial bank expansion systems, with micro-level dense bank branches and numerous salespeople. For the previous generation of algorithmic stablecoins, they aimed to achieve credit expansion through a complete set of smart contracts and internet front-end operations. However, this article argues that relying solely on smart contracts for stablecoins is problematic; a retail model is still needed, as some operations must be conducted by people. Humans have subjective initiative, and more importantly, everyone has strong information specificity regarding their assets, which forms the basis for credit expansion.

Marginal Issues in Credit Expansion

Interest rates should change at the margin within a certain range; otherwise, a series of problems will arise. Algorithmic stablecoins relying on smart contracts or specific protocols for credit expansion treat all participants equally regarding entry thresholds and collateral ratios. This creates a problem of adverse selection. When a person considers borrowing or collateralizing assets to obtain stablecoins, if they find the collateral ratio unsuitable, they will choose what they perceive to be a better project, most likely one where they believe they are getting a deal.

Therefore, if marginal conditions are uniform, the users choosing this project will be worse than expected, leading to serious adverse selection issues for the entire project, meaning the probability of default is higher than anticipated. This is why adjustments must be made at the margin for individuals. This has been repeatedly validated in both traditional finance and decentralized finance. To solve this problem, it is necessary to introduce a credit retail model.

Credit Retail Model

This model requires a complete on-chain credit system, where overall credit expansion relies on salespeople's credit issuance at a micro level. Salespeople work for themselves while also providing some guarantee for credit. They can freely form organizations and determine their levels and quota discounts based on their historical performance and collateral in the stablecoin project.

The project party is only responsible for ensuring payment settlement and credit settlement, as well as maintaining a fixed yield market. Salespeople form initial credit based on historical performance and collateral, obtaining limits and interest rate ranges, and then lend based on market loan demand. Borrowers need to provide credit records and collateralize part of their assets. This way, currency is issued at a micro level, achieving credit expansion.

The advantage of the retail model is that it can leverage human information advantages at the margin. By constructing a credit expansion system through Web3, if DID develops well, users can use part of their information as collateral. This process effectively captures external commercial credit, as currency creates incremental value during this process. Liquidity is not allocated to arbitrageurs but to those with genuine borrowing needs, and these needs constitute commercial credit. As commercial credit is continuously realized, the entire system gains internal growth momentum.

Building a Monetary Policy Regulation Framework

Native Token Design

In previous projects, most algorithmic stablecoin projects issuing native tokens took all the seigniorage, motivated by the expectation of currency issuance to incentivize people to buy native tokens, which in turn supports stablecoins. However, this form of credit has proven to be fragile. The unreasonable aspect of taking all the seigniorage is that it should not be controlled by a few entities, nor should it incentivize a few people to buy large amounts of native tokens.

The formation of the entire credit system cannot be achieved by a single team; it relies on all participants within the system. Unreasonable distribution of native tokens can lead to imbalances in the benefits brought by system expansion, with a few individuals rapidly accumulating large amounts of wealth in a short time, which is a failure for the incentive system. Benefits should be distributed equitably, and rules should be clear to slow down wealth distribution, which is suitable for the overall development of the system. Therefore, this article proposes that native tokens should absorb part of the seigniorage.

Thus, native tokens should be issued as part of the reserve assets to absorb some seigniorage, and the proportion should increase with credit expansion. The proportion of native assets as reserve assets reflects the degree to which this form of credit is accepted by the public. The design of native tokens should follow these points:

• Native tokens should only take a small portion of seigniorage; at this stage, stability and decentralization should be prioritized.
• As the internal yield of the system increases, gradually transition to using future income from native tokens as reserve assets to support currency issuance, with the proportion slowly increasing.
• Flexibly choose cycles; during accelerated expansion, the proportion of native tokens can be appropriately increased.

Monetary Policy Quantity-Price Regulation

Quantity refers to the money supply, while price refers to the yield level of the entire system, directly reflecting the liquidity price at the margin. Directly regulating the money supply is the most straightforward monetary policy. The previous generation's monetary policy regulation mechanisms, such as rebase, are consistent at the margin and do not involve precise allocation. If a credit retail model is used to issue currency, direct adjustments to the quotas of salespeople can be made to directly inject currency at the margin.

Compared to directly regulating the money supply, price regulation is more precise. This requires a well-developed liquidity trading market to form a yield, while the project party has a controllable interest rate as a benchmark rate to influence the entire market. For projects expanding using a credit retail model, the benchmark interest rate can be based on the collateral ratio.

In price regulation, a counter-cyclical adjustment mechanism can be used to guide price regulation.

• During accelerated expansion, when money demand is high and currency issuance is large, increase interest rates (collateral ratios).
• During accelerated recession, when money demand is low and currency issuance is small, lower interest rates (collateral ratios).

Through counter-cyclical adjustments, the project's volatility can be reduced, but specific operations must still be chosen based on cyclical patterns and strategies.

Conclusion

Stablecoin projects are constrained by the impossible triangle in the long term, so it is essential to have a clear understanding of the impossible triangle and to achieve dynamic balance according to dynamic rules.

If credit expansion relies entirely on contracts, there will be adverse selection issues, and there will be no changes at the margin. Therefore, adopting a credit retail model, relying on salespeople's lending to capture external commercial credit, achieves credit expansion.

The acceptance of the credit form of native tokens is a long-term process. Therefore, native tokens can only absorb part of the seigniorage. In terms of quantity-price regulation, based on the credit retail model, one can directly adjust the quotas of salespeople to regulate the money supply and adjust liquidity prices through collateral ratios, thereby influencing the liquidity market.