DeFi Risk Removal: Analyzing Systemic Risk in Decentralized Systems

There is never a one-size-fits-all way to avoid financial disaster.

In an unpredictable society, systemic risk is only partially addressed. Economic conditions, technology, and human behavior all change over time, so the approach to systemic risk must also remain variable. This development will not necessarily lead to a more efficient and stable state, as it will be continuously influenced by innovation, regulatory actions, changing psychological patterns and behaviors of financial market participants. This is true in both traditional finance and decentralized finance.

We define systemic risk as: risk in an interconnected network of agents where pain caused by one or more agents can spread to several other agents in the network, resulting in a widespread crisis. Systemic risk has traditionally been associated with institutional failure. Of these, defaults can lead to a catalytic increase in the cost of capital, an example being a bank run, which forces a bank into bankruptcy. Because banks often lend to each other, defaults can trigger a domino effect.

Generally speaking, there are 3 situations that are considered systemic risk.

  • Information contagion – depositors’ anticipation of the possibility of a crisis leads to a bank run.


  • Direct contagion – spread through financial linkages such as debt/credit relationships created by the interbank level and system, or other exposures such as intercompany credit chains.


  • Common risk – a decline in the value of an asset, the risk is to agents holding the same or related assets.


contagion graph theory

Let us analyze the impact of default contagion from a graph theory perspective. If the extra links help distribute a loss flow proportional to their absorption capacity among the nodes in the network, the network will become more robust in terms of resistance to contagion. The beneficial effect of connectivity can only be guaranteed with a slight special shock under the following very strict requirements:

1) The loss flow spreads out over the network N, along the directed tree

2) All nodes have the same absorption capacity and outward expansion capacity.

DeFi Risk Removal: Analyzing Systemic Risk in Decentralized Systems

In the image above, the specific shock hits the start node of the dashed line. The addition of links does some damage to previously untouched parts of the web. Therefore, the burden of loss is borne by more nodes, thereby reducing the impact of traffic on each individual node. However, if the impact is too great, the increased connectivity can lead to cascade failures.

In closed paths and loops, the network produces uneven loss distribution.

DeFi Risk Removal: Analyzing Systemic Risk in Decentralized Systems

closed path case

The dashed line connecting B to E creates a closed path, allowing E to carry losses while reducing traffic that might go to C and D. If C, D, and E have the same absorption capacity, the action of the dotted line allows the system to better handle the infection.

The same reason applies to circular paths. In the figure below, the dashed line creates a cycle of A => B => D => A and the loss is reduced to E.

DeFi Risk Removal: Analyzing Systemic Risk in Decentralized Systems

Circular path situation

Therefore, networks with low connectivity are more likely to be tree-like networks, while networks with high connectivity are more likely to be loops and closed paths. Higher degrees of connectivity lead to more diverse lending behaviors. Adding links to a network with high connectivity increases the number of loops and closed paths, thus reducing the benefits of decentralized diversity. Thus, when the connectivity of the network is at its peak, the benefits of diversification are also maximized.

Based on the above situation, we can propose a theory:

In networks where the ratio of the number of directed links to the number of nodes is low, connections will increase. The network structure that is most resilient to contagion is the one with the highest connection probability.

DeFi Risk Removal: Analyzing Systemic Risk in Decentralized Systems

Shows the impact of advanced networks on the biggest shocks experienced by banks

To demonstrate the relationship between higher-order networks and experienced min/max shocks, we use traditional banking networks to simulate market shocks and contagion events, inspired by Stuart Gordon Reid. We created an interbank network model with about 50 banks. Then we uniformly generate a random number between 50 and 2500 to represent the number of connections in the network (nodes are randomly selected). When the number of connections equals 50, each bank is only connected to one other bank. In theory, when the number of connections equals 2500, all banks are interconnected.

After the network is created, simulate the shock to the system and observe the contagion spread. As the shock propagates through the network, the impact of the shock is proportional to the number of neighbors connected to each bank.

These charts help visualize trends and have been smoothed. From this preliminary experiment, we can see that an increase in network size is positively correlated with an overall increase in network stability, but after a certain point, stability deteriorates. This is fairly intuitive, as fully connected networks are most vulnerable to systemic risk contagion. Here are some aspects that can also be translated into a crypto network, where each validator node represents each bank node, although the model is very simple.

DeFi Risk Removal: Analyzing Systemic Risk in Decentralized Systems

Shows the impact of higher-order networks on the smallest shock experienced by the bank

Systemic Risk in the Blockchain Environment

So far, we have only discussed the systemic risks of generalized node networks, which can be attributed to various networks, macro and micro, etc. However, systemic risk only increases as the system becomes more complex. Blockchain technology has fundamentally changed the market structure of derivatives. By contrast, central counterparties create risk by creating large entities that are prone to failure. The decentralized clearing function of the blockchain can reduce the risks caused by excessive centralization. The ideal blockchain-based system decentralizes the clearing function and distributes these tasks to network members without creating unequal pressure.

Let us consider a decentralized clearing system that mitigates systemic risk by default contagion. The main challenges of the actual clearing function are determining available funds and resolving disagreements over payment seniority. In a blockchain clearing mechanism, these two issues are resolved automatically, rather than through an intermediary, reducing friction. Thus, the blockchain system both improves the recovery rate of defaulted assets and increases the bank’s liability for risky transactions.

Alas, we find ourselves repeating an old adage: it is not gold that shines. Due to the unavoidable need for a centralized form of governance, and the tendency for consensus mechanisms to have concrete powers, blockchain systems have only an illusion of decentralization. DeFi in particular exhibits many vulnerabilities due to high leverage and liquidity mismatches. In some applications, the built-in interconnectivity has a high probability of cascading failures in the event of a shock.

DeFi Risk Removal: Analyzing Systemic Risk in Decentralized Systems

The main difference between DeFi and CeFi in the crypto space is whether financial services are automated through smart contracts or handled by centralized intermediaries and stablecoin designs

Systemic Risk Mitigation

The oracle system is the mouthpiece of the smart contract, which is often cited as a critical point of failure. A case study that best examines this claim can be found in the popular decentralized oracle service Chainlink.

Chainlink effectively applies our theory to risk-efficient decentralized systems through a three-pronged approach:

  1. Distributed data source
  2. Distributed Oracle
  3. Use trusted hardware.


We will now take a closer look at how a distributed oracle can help make it a contagious risk-tolerant system. 

Chainlink builds a modular rather than monolithic system to ensure risk is not concentrated in one oracle. This will create a collection of n distinct oracle nodes { O 1 , O 2 , …, O n } , each oracle contacting its own distinct dataset source.

One of Chainlink’s best ways to prevent cascading failures is to prevent oracles from replicating each other. Imagine an oracle Oz observing the response of another oracle Oi and replicating it. This can lead to weaker security by reducing the diversity of data sources, which can lead to erroneous responses across the system. Chainlink avoids this with the commit/show algorithm. The algorithm below shows a protocol that guarantees availability, given 3 f + 1 nodes. Only after all commits are made, the Oracle response is resubmitted and exposed to potential replication. This precludes a cheating oracle from duplicating another oracle’s response.

DeFi Risk Removal: Analyzing Systemic Risk in Decentralized Systems

Source: chainlink white paper

Given a total of 3f + 1 nodes, at most f will be defective, which means that at least 2 f + 1 will send commitments in step 4. At most f of these commitments comes from the node in question, so at least f + 1 comes from the trusted node. Since at least one f+1 commitment on a single value A must come from an honest node, it is clear that A or the aggregated response will be accurate as the result of the algorithm.

It is difficult to reach consensus on the value A while recognizing that there may be faulty nodes, which is similar to the Byzantine generals problem.

Lack of consensus is a systemic risk

The Byzantine Fault Tolerant (BFT) consensus protocol is a Byzantine solution.

The Byzantine Generals Problem, proposed by Lamport, Shostak, and Pease in 1982. The question states that there is an army scattered around the city consisting of a general and n-1 lieutenants. The army is preparing to attack a common enemy, but has not yet decided when to do so. The attack will only succeed if the entire army charges at the same time. By sending signals back and forth, the general and his lieutenants must agree on the best timing to strike. However, some lieutenants are traitors, which means they can lie about their decisions.

The Byzantine General’s problem is similar to the blockchain problem, where the network (the general and his lieutenant) must agree on broadcasting transactions (the timing of the attack) even if some nodes are unreliable (the lieutenant betrays). Byzantine fault tolerance is a property of a system that allows a given number of failures from the Byzantine Generals problem to be tolerated, thereby reducing the risk of communication failures.

The traditional financial system is not a BFT system. TradFi still falls prey to dishonesty or misinformation input, and as a result, disaster ensues. The problem is so common that a crack in the wall could end up escalating into the destruction of the entire house, so to speak.

How to secure the future

Improving the underlying infrastructure of DeFi will require extensive efforts by public regulators and private sector actors. In the past, regulation has tended to combine the two, with periods of more comprehensive government regulation stemming from systemic crises, or the failure of the private sector to enforce self-imposed standards.

In general, the more effective self-regulation is at limiting crises and protecting customers, the harder it will be for state regulation to gain political support. As such, the overall DeFi regulatory framework will be significantly influenced by the type and effectiveness of private sector self-regulatory initiatives. Risk insurers, DeFi service providers, and end users will all benefit from the increased level of system security, which creates a lot of incentives. Insurance providers have already contributed to guiding smart contract security best practices, however, there is always room for thoughtful experimentation, developing insurance products, developing standards and technologies to reduce the likelihood of catastrophic events and their collateral damage. is worth it.

Finally, a higher degree of digitization, transparency, automation, and Byzantine fault tolerance are the main technical advantages of DeFi in mitigating systemic risks compared to traditional finance . Decentralizing the clearing function and distributing these chores among network members in a proportional manner (ie: directed links ≤ nodes) is an ideal blockchain-based risk mitigation solution. The more open source code and publicly verifiable ledgers are used in DeFi, the easier it will be to build automated risk simulations, stress testing, monitoring, early warning signals, circuit breakers, insurance coverage, claims processing, reporting, and other integrated forms of risk management.

Posted by:CoinYuppie,Reprinted with attribution to:
Coinyuppie is an open information publishing platform, all information provided is not related to the views and positions of coinyuppie, and does not constitute any investment and financial advice. Users are expected to carefully screen and prevent risks.

Like (0)
Donate Buy me a coffee Buy me a coffee
Previous 2022-06-07 23:42
Next 2022-06-07 23:43

Related articles