Tesla changes mind on BTC, Binance under investigation, Grayscale hits new low
GME saga highlights settlement woes blockchain tech could potentially solve
A discussion around blockchain tech's potential to replace the archaic T+2 settlement system
The Wall Street Bets movement and the ensuing influx of money into the heavily shorted GameStop (GME) stock have been one of the headline stories of 2021. Retail traders inspired by the movement poured money into GME, driving its price up by almost 8,000% and forcing a "short squeeze" that hurt the institutional investors who had bet against the company.
The stock, trading around $19 on Jan. 4, went on to peak at $465 on Jan. 28, before one of the most popular commission-free brokerage services among retail investors, Robinhood, barred users from opening long positions on the stocks of GameStop, telecommunications company Nokia and the world's largest cinema chain, AMC.
So, how did a company with a dying, outdated business model become the most talked-about stock? And what prompted a popular brokerage to restrict trading despite the massive demand?
At first glance, it is a milestone story about "evil" hedge funds vs. retail traders who united on Reddit to fight the big guys on Wall Street. Since hedge funds had heavily shorted GameStop, retail traders bought up the company's shares, raising its value and forcing hedge funds to close their short positions at a heavy loss. One of the biggest among them, Melvin Capital, lost nearly half of its capital.
Soon after, brokerage companies such as Robinhood — which had temporarily limited the purchase of GameStop shares for their retail clients — were accused of helping the sinking hedge funds, and class-action suits were filed against them.
While this story touches many ideological and political issues, at its core, it highlights a deeper, more technical issue: inefficient settlement.
The archaic, inefficient settlement system of today
While the popular narrative is to blame brokers for siding with hedge funds, the need to stop the buying of certain shares stems from the archaic settlement system that is the backbone of the U.S. financial market.
In order to provide a more accessible experience to clients, brokers show them their entire positions as they enter or exit them. However, in reality, all brokers operate on a "T+2" scheme. According to this system, when a broker buys shares for a client, they are only settled, or received, after two business days, and the money against them is also paid at that time.
This means brokers must settle accounts with the Depository Trust & Clearing Corp., or DTCC, against their clients' positions in two days, at the original price. As the spectacular rise in prices of shares such as GameStop continued, the DTCC demanded large sums of collateral from brokerages, including Robinhood.
Clearinghouses such as the DTCC are just intermediaries between buyers and sellers. Their purpose is to protect market participants by ensuring that parties like brokerage firms have enough funds to back up the trades they make on behalf of their clients. Because the settlement of trades happens two days after the transactions (T+2), there’s a risk that the broker won’t be able to pay for shares at the time of settlement. Additionally, in certain high-risk cases (e.g., GameStop), there is a real possibility that the stock could lose value very rapidly.
Hence, in response to the DTCC's demands for higher collateral, Robinhood had to borrow $1 billion at first, and then another $2.4 billion. As a result of this crunch, the brokerage had to restrict certain trades, resulting in public criticism and outcry from customers.
Naturally, trading restrictions adversely affected everyone but the hedge funds. However, it likely had more to do with the T+2 settlement system, as opposed to a larger conspiracy at play.
This delayed settlement process is an old mechanism — and only around four years ago, the securities markets moved from a T+3 to a T+2 settlement period. However, since then, millions of new retail investors have entered the market as online brokerages transformed technology in this space. Given the current demands on trading platforms, the T+2 settlement is far from ideal.
Speaking on the matter, the CEO of Robinhood Vlad Tenev said on Feb. 5, “There is no reason why the greatest financial system the world has ever seen cannot settle trades in real-time.”
Later, on Feb. 18, there was a hearing on the Robinhood case in Congress, and the brokerage provided a detailed explanation of what had transpired. Notably, the current settlement process and its shortcomings were highlighted.
While there are financial and technological challenges when it comes to changing the current settlement system, in this article, we consider whether a blockchain-based system could prove useful in such a scenario.
Real-time settlement on blockchain
Many market participants consider blockchain as a way to seriously transform securities settlement. The major innovation would be to build a public securities database with ownership records so that it can be maintained without any intermediaries or third-party infrastructure. Assets would then be settled through a system of smart contracts in the absence of a central authority.
In such a model, Participant A sends a settlement instruction in the form of a smart contract, and Participant B responds with a matching settlement instruction. The smart contract would hold all the relevant information — i.e., specifying the assets involved, their quantity, the blockchain addresses of both participants, and more.
In this case, all functions that are currently conducted by clearinghouses like DTCC would be performed by a computer program without the need for reconciliation and delayed settlement.
Despite the many challenges, including capacity constraints, some projects are preparing to present blockchain-based settlement solutions soon. Their designs support market data, peer-to-peer trading limits, execution of orders as well as the calculation of the net aggregated positions per asset, per counterparty.
That being said, major clearinghouses and financial institutions also realize the opportunity at hand, as reflected in the R3 CEV consortium and the Linq project by Nasdaq. The DTCC has also launched a project called "Ion" to provide a new framework for asset digitalization and tokenization. Ion is DTCC's initiative aimed at improving post-trade settlement in public and private markets.
As part of the Ion project, DTCC envisions a digital settlement service that combines the benefits of accelerated settlement but keeps the advantages of central authority to prevent fragmentation of the clearing and settlement space. DTCC has already developed a proof of concept and is evaluating market feedback for the initiative.
In 2019, the Division of Trading and Markets of the U.S. Securities and Exchange Commission granted no-action relief to Paxos Trust Company to conduct a “feasibility study” of a securities settlement service using blockchain technology. During this period, Paxos can operate as a clearing agency under Section 3(a)(23) of the Securities Exchange Act without needing to register as a clearing agency under Section 17A(b)(1) of the Act.
While these initiatives are a major step forward, they are yet to materialize and come with their own sets of limitations and risks.
Limitations of real-time settlement on the blockchain
It is clear that blockchain technology could potentially solve the delayed assets settlement problem, but there are several concerns that need to be addressed before actual implementation.
Inequality and scalability risks
In the case of securities settlement, the blockchain is just a record-keeping system that tracks securities ownership and payments related to securities trades. The general public sees blockchain as a genuine fairness instrument because it is permissionless and there are no third parties that can arbitrarily update the information stored on it, so every participant is in the same position.
However, it is not entirely impossible to manipulate blockchain records to some extent. For example, if a blockchain achieves consensus (i.e., an agreed-upon state) based on a proof-of-work protocol, a group of transaction validators, called miners, is required to solve computationally intensive problems. The mining entity that solves the problem updates the blockchain and obtains a reward.
Moreover, a blockchain consists of blocks that store the records of transactions in batches. The block time of a blockchain defines the frequency of these batches, and users can accelerate their transaction's settlement time by posting a higher transaction fee, which incentivizes miners to confirm their transaction over others.
This essentially means that whoever can afford paying higher fees receives faster asset settlement, especially when the blockchain is congested during peak usage. In the long run, this mechanism runs the risk of turning blockchain from an equitable technology to one that services only the elite. To put this into perspective, in times of high volatility and usage, the number one cryptocurrency, Bitcoin, can take more than 24 hours to confirm transactions — a period not significantly different from the T+2 settlement scheme.
Another example of such operational considerations is the so-called "Transaction Ordering Dependency" with smart contracts, which is on display in the DeFi space. This happens if two dependent transactions use the same smart contract and are in the same transaction block. In this scenario, since miners/validators prioritize transactions with the highest fees, an entity can potentially front-run another by copying their transaction and posting it in the same block with a higher fee attached.
In the DeFi space, participants regularly scan the blockchain for arbitrage opportunities and then copy and paste the same trades with higher transaction fees in order to outbid the original trader.
If we lend credence to the conspiracy theories around hedge funds stifling retail traders, a potential shift to a blockchain-based system could, in theory, give wealthier market participants the ability to use the system's transparency to front-run individual traders.
Finally, if asset settlement is expensive in terms of fees, or very slow, a lot of participants might choose not to commit to a transaction that will make trading ineffective. Thus, the feasibility of such a blockchain depends on whether the system can provide the settlement in a cost-effective and timely manner.
When asset settlement happens on a blockchain, it is nearly instantaneous, and while the increase in speed is what we are aiming for, other risks emerge in such a scenario. For instance, the quick settlement would be detrimental to an entity whose securities have been stolen. In traditional markets, the delay between settlement allows time to take legal action, but on the blockchain, such securities could be sold in minutes.
Another example of this pertains to the limited amount of time on a blockchain system to confirm or cancel transactions — and during times of congestion, this can become a major limitation.
MakerDAO is one of the largest DeFi protocols and uses smart contracts to generate and maintain DAI, a stablecoin soft-pegged to the U.S. dollar. The minting of DAI happens through several “collateralized debt positions" that require Ether (ETH) as collateral. In March 2020, on Black Thursday, the price of Ethereum's native coin dropped 43% in just a few hours. As a result, MakerDAO saw many CDPs suddenly liquidated due to the value of their ETH collateral dropping below the minimum threshold required at the time.
Even though MakerDAO gives CDP owners the option to add more collateral to their positions, during the “Black Thursday'' crash on March 12, 2020, CDP owners could not access their vault fast enough to add more ETH. Moreover, high gas prices due to congestion also caused a signiﬁcant lag in updating the price of ETH collaterals with oracles, resulting in many CDPs being liquidated once the price was eventually updated. According to the research team whiterabbit, 1,462 of the 3,994 liquidated CDPs, or 36.6%, were liquidated with a 100% discount.
In the situation with GameStop's stock, given its high volatility, real-time asset settlement might also have led to some unexpected losses — especially, if other parts of the market infrastructure had any weaknesses or, as was the case with GME, if there were heavy use of derivatives, as many retail traders used margin and options to trade GameStop shares.
A more in-depth discussion on issues surrounding counterparty and timing risks can be found in this paper.
When we consider asset settlement on a blockchain, or through a system of smart contracts, one of the most critical risks arises in the technical side of the system. There is a real risk that bad actors could exploit code vulnerabilities, and there are numerous recent examples of security breaches within the DeFi ecosystem.
One of the major smart contract exploits was the famous "DAO hack," which happened when a hacker used a "re-entrancy" vulnerability to steal $60 million worth of ETH from a smart contract. Given that smart contracts are designed to directly hold and release funds automatically, they are very attractive targets for hackers.
Since smart contracts are essentially just sets of executable code that run on the blockchain and enforce predefined terms, they are prone to numerous vulnerabilities, some of which are discussed below.
Re-entrancy: This hack causes a contract to send funds before it has updated its internal state. Essentially, it calls the function to request funds from the contract again and again, thereby draining it. This vulnerability is also highlighted in the official Solidity documentation and was used in the DAO hack.
Integer underﬂow/overﬂow: This is an old and widespread vulnerability that is not specific to smart contracts. In programming languages, an integer is just a whole number, and when a program receives a number that is too large or too low, it can cause abnormal behavior.
Smart contracts aren’t an exception to this, and one study found the most common vulnerability in smart contracts to be "Integer Overﬂow." A more detailed survey on this vulnerability in relation to smart contracts is available here.
Timestamp dependence: This is relevant to contracts where the block timestamp is a condition for transaction execution. Here, the contract output can potentially be manipulated by a miner/validator because the timestamp is defined by the time on the miner/validator's local computer or server. This presents an opportunity for a bad actor to choose different block timestamps and change the outcome of a smart contract that has a timestamp-dependent output.
All modern blockchains only have on-chain data, which means that, by default, the transactions on the blockchain aren’t connected to any real-world data — such as the prices of other assets, data from exchanges, etc. For asset trading and settlement to be facilitated on the blockchain, there is a need for software infrastructure to link traditional finance with blockchains and blockchains with each other.
Given that this infrastructure is providing the data feeds that are later used in smart contracts and transaction execution, it should be secure and reliable around the clock. Problems like delays in the data feed, power/internet outages and hardware failures might cause incorrect smart contract outputs and result in large losses for stakeholders.
Currently, some of the most popular tools for blockchain data feeds are so-called oracles. These oracles are not data sources themselves but merely relays that provide a variety of external data to blockchains. The major risks presented by such a system include oracles' inability to receive and relay data in a timely manner — and in the worst-case scenario, bad actors manipulating oracles to relay tampered data.
Currently, nearly all existing blockchain-based asset settlement systems in the DeFi space are reliant on oracles. Many of these oracles are centralized and mainly utilize trust models, while the design of the blockchain is trustless. Clearly, oracles are far from an ideal solution, and their mechanics are still vague and unexplored.
Apart from these operational risks, it is pertinent to mention that blockchain-based systems have no support centers, and transactions cannot be reversed at will. This means the need for security, reliability and risk-reduction is greater than ever and must be in place before we advocate such an implementation.
As discussed, while the proposition of replacing the archaic T+2 settlement system with a blockchain solution is exciting, the current state of the technology and infrastructure is not ideal for such a transition.
That being said, the blockchain space is developing extremely quickly, and innovation in the sector is incredible. Given the technology's potential to dramatically change existing systems, it would come as no surprise if — once its limitations are addressed — we see the financial systems of the future utilizing blockchains.
Finally, regulators and policymakers who understand the need for this type of system, and who are willing to facilitate it, are critical for such a change. A decisive shift to central bank-issued digital currencies, for instance, would be a big step toward such a transition.
This report was written by the Financial Markets team at OKEx.