Key Insights

  • The Vega Protocol, currently in restricted mainnet, is a standalone proof-of-stake blockchain with network architecture optimized for trading derivatives.
  • Vega is built on the Tendermint consensus algorithm and reconciles the lack of native network assets with an Ethereum bridge, allowing participants to use any Ethereum-based ERC-20 asset on-chain.
  • Permissionless market creation allows users to launch all sorts of new markets, in terms of product, asset, and settlement, by customizing modules at different layers of abstraction.
  • Unique incentives and precise liquidity pricing along with the option for market makers to be active or passive helps bootstrap liquidity for nascent markets and provide active, competitive pricing and volume where it is needed most.
  • Vega’s risk models ensure market and protocol solvency while using technical features to retain capital efficiency.

On March 8th there was fury across the global trading community after the London Metals Exchange (LME) canceled $4 billion in nickel trades. The LME, owned by Hong Kong Exchanges and Clearing, retroactively reversed these trades because China’s Tsingshan Holding Group faced billions of dollars in losses.

This fiasco was effectively an advertisement for the Vega Protocol, a decentralized network for margined financial products. As a sovereign blockchain designed specifically for trading derivatives, Vega has developed novel approaches to fairness, risk monitoring and capital efficiency, while mirroring a centralized experience with its high performance. Also, beyond just preventing an event like the LME nickel debacle, Vega is tearing down the high barriers in traditional finance and much of crypto to enable permissionless, customizable market creation.

Credible neutrality, risk management and open access are crucial for the $600 trillion (notional value) derivatives market. Although commonly written off as an outlet for financial speculation, derivatives play a vital role in economic growth and development. The 145-year-old London Metals Exchange, for example, is the largest derivatives market for base metals and helps metal producers and users mitigate the cash-flow volatility tied to price risk. In order for these important markets to remain useful and liquid, they must be functional and fair.

Vega Protocol founder Barney Mannerings, a computer scientist with 15 years of capital markets experience, saw the blockchain’s role in building a better, more open financial system early on. But, having designed multiple releases of the London Stock Exchange’s core trading system and matching engine, he didn’t consider Ethereum’s performance and other on-chain properties to be suitable for complex, high-performance trading. However, in early 2018, he realized he could build a well-functioning, decentralized trading network from scratch.

Network Architecture

Today, Vega is a standalone, custom-built, proof-of-stake blockchain that runs on the Tendermint consensus mechanism. Vega’s network architecture is optimized for trading on several dimensions, starting with speed. Sophisticated trading systems need high throughput and low latency to support native order books and ensure that orders are priced correctly. Vega’s consensus layer achieves this with its 1 second block time and ability to process thousands of transactions per block, with instant finality.

The architecture’s second optimization deals with fees. Vega’s mainnet will offer continuous trading through either an on-chain central limit order book (CLOB) or a hybrid design using both an automated market maker (AMM) and a CLOB. In this system, limit orders are valuable because they provide information and liquidity to the market. While generic blockchains charge fees based on the computing power needed to process a transaction, Vega wants to incentivize limit orders, so there is no fee to place orders on the Vega chain. Fees are only charged when orders execute. To prevent order spamming, Vega has implemented client-side proof-of-work for transactions, so as a user starts putting more orders into a block it becomes more expensive (in terms of effort in the proof of work) for a transaction to be accepted into the mempool.

Another issue facing DeFi is that entities such as miners often use the transparency of the blockchain to front-run profitable transactions. To neutralize this Vega designed an ordering protocol called Wendy. Wendy runs alongside consensus and uses the same trust assumptions about validators as proof-of-stake. So, if at least two-thirds of validators are honest about when they see messages, users can be assured that transactions will be ordered fairly.

Markets

On top of the Vega network lie its markets. In traditional finance, market makers and exchanges, such as Goldman Sachs and the CME Group, are gatekeepers that decide the ‘who’ and ‘what’ for market creation and trading. Vega disintermediates this process and allows anyone to create open (public) markets using customizable modules. This creation process is relatively accessible thanks to Vega’s WebAssembly compiler, which lets participants build instruments and smart products with just about any high-level programming language. Since Vega markets are layered, prospective architects will start at the base, the product layer.

Another way to visualize markets on Vega. Source: Vega Whitepaper

The product layer allows anyone to pseudonymously create any type of market. Users can set up more traditional markets, such as spot swaps, futures, options, and perpetual swaps, or design new markets altogether, as long as they specify how to value the product and calculate settlement flows. The product creator must also specify when to calculate settlement, which could be as cash flows are generated (recurring), when positions are closed, or at an expiry date. Another important piece of product logic is the trading mode, which can range from open continuous trading through a CLOB to private, closed, over-the-counter (OTC) trades via off-chain matching. Finally, the user specifies the parameters (i.e., strike price, date of expiry, maximum leverage, etc.).

On top of the product layer, the next layer of abstraction that users can build on is the instrument layer. An instrument is just a product with filled parameters. An example is filling in the parameters for a generic call option product to create a BTC call option instrument; e.g., reference to a pre-approved oracle for the price of BTC, a strike price of $100,000, and an expiration date of 12/31/2022.

Finally, to make the instrument tradable, it must be combined with one of Vega’s risk models and appropriate values for the risk model’s parameters. The risk model’s main job is to determine the collateralization requirement for open positions. Once a tradable instrument has been proposed for open market listing, it will be voted on by Vega’s on-chain governance.

Liquidity and Market Making

Managing liquidity is crucial for decentralized exchanges since insufficient liquidity makes it difficult to execute trades without significant price impact. Vega attempts to attract liquidity from different angles with the ultimate goal of providing valuable liquidity, meaning liquidity that trades.

Liquidity pricing is based on open interest, trading volume, and current liquidity (i.e., order book depth), among other factors. The objective is to price liquidity so that it flows to markets that have the greatest need.

Liquidity Supply vs. Cost of Liquidity. Source: Vega Whitepaper

Vega also uses an owner-operator model for market makers to steer liquidity to markets that need it. Market makers receive an equity-like stake in trading fees, proportional to the trading volume in a market when they added liquidity. Thus, market makers are incentivized to seed or create good markets and provide long-term support.

Because most open markets on Vega will trade continuously through the hybrid AMM-CLOB, a market maker can choose to either be active or passive. Active market makers, as the name suggests, actively manage pricing for their volume and must do so within a specified band around the last traded price. Passive market makers have their capital deployed automatically and algorithmically based on market prices. Even though CLOBs are much better than AMMs for trading derivatives, they are considerably worse for bootstrapping liquidity, so the AMM is an important complement in the early and illiquid phases of a market.

Both active and passive market makers will earn trading fees in exchange for providing liquidity. Trading fees will initially be bundled into a reward pool and then distributed to individual market makers based on their equity-like market share (stake in the trading fee pool), activity (active market makers with competitive prices earn more), and length of their market-making commitment. In general, fees will come from the price (liquidity) taker and be distributed to the price maker in the particular trade, infrastructure providers (validators), and, as outlined above, the market maker reward pool.

Anyone can become a market maker as long as they provide a financial stake, using the traded assets, as a bond. If they fail to provide active volume in proportion to their financial stake, they are penalized and lose part of the stake to the insurance pool. This mechanism helps prevent liquidity shocks.

Trading

Irrespective of liquidity, new markets start in an auction setting to discover the initial price, before moving to their normal trading mode. During regular trading, protective auctions are a backup, only run when circuit breakers are triggered. These protective auctions start when a market crosses the predefined threshold (as defined by its risk model) for price volatility or low liquidity. Under such circumstances, auctions let participants figure out a fair market price or source liquidity.

Once an open market does find its initial price, it starts its normal trading mode. Continuous trading will either use a standard CLOB or a hybrid AMM-CLOB. Discrete trading translates into frequent batch auctions, introduced as a way to neutralize high-frequency trading.

Less common but still useful are ad-hoc markets, which are closed (private), and usually created for one-off use. Commonly referred to as over-the-counter (OTC), in these trades, a party can advertise a trade and select the best bid (called a request for quote) or two parties can agree to a price off-chain (called a matched trade) and use the Vega network for execution and settlement.

Collateral and Settlement

The first requirement when trading a margined financial instrument is to have sufficient collateral. This is especially important for Vega since all market participants operate under pseudonymity, so the protocol cannot gauge creditworthiness or have recourse. Since Vega doesn’t have any native network assets, it needs to source collateral and trading capital from other chains. Even though Vega’s use of Tendermint makes it a part of the Cosmos ecosystem, it was designed before the Cosmos SDK or Inter-Blockchain Communication Protocol (IBC) and has not bridged to the Cosmos Hub due to the overhead from network requirements. Instead, Vega has its own Ethereum-based ERC-20 bridge, since Ethereum still holds the majority of liquid, tradable value. Any ERC-20 can be used as a market’s base currency (for pricing, settlement, and margin).

Once a user has deposited collateral to Vega from the host chain they can place an order, which will reserve a certain amount of margin. If the order trades and a position is generated, Vega will start monitoring that position through its collateral manager module. Since there is no recourse if a trader owes more than their collateral, positions must be resolved while a trader’s margin is greater than their loss. Hence, risk parameters such as initial margin (to open a position) and maintenance margin (to keep a position open) are stringent. However, rather than just bloating margin requirements, Vega uses live position monitoring, continuous risk recalculation, cross-margining, and (opt-in) automatic recapitalization to bolster capital efficiency.

Continuous monitoring lowers margin costs because it compresses the time over which a position can move against a trader, so Vega ends up needing less collateral. Constant P&L monitoring also allows for cross margining, meaning that positions in profit can release some margin to offset losses and net margin across markets without having to close positions. The settlement engine also automatically sweeps funds back into a user’s wallet once a position closes, and if a user opts in, it can also automatically sweep funds from a wallet into a position approaching its maintenance margin.

As soon as a trader’s available collateral falls below the maintenance margin, the position is closed out. Referred to as “atomic closeouts,” these instant liquidations by the protocol prevent further price movement against the position. Liquidations are designed so that positions are closed out with a net positive margin. This remaining sum is then added to the market’s insurance pool to cover any position closed out with a negative balance.

The VEGA Token

VEGA is the Vega Protocol’s Ethereum-based, ERC-20 staking and governance token.

Vega is a delegated proof-of-stake blockchain, which means that the validator nodes running the network need to have, at least, a certain amount of VEGA staked. However, participants holding any balance of VEGA can delegate their stake to validator nodes. Stakers (validators and delegating token holders) earn a share of trading fees for securing and running the network, which means the fundamental value of VEGA will increase with the protocol’s trading volume.

Unlike most blockchains, traders on Vega are not required to buy and use VEGA to access the protocol since there are no gas fees.

VEGA also allows tokenholders, whether they have delegated their tokens or not, to create and vote on governance proposals. Governance proposals cover the creation of new markets, the alteration of parameters in active markets, or changes in network (chain) parameters (e.g., the infrastructure fee given to stakers).

Most token allocations have already started vesting and will vest linearly into late 2023. Locked tokenholders can still stake and vote.

Up Next for Vega

Currently, there are two ways to use Vega: through the restricted mainnet or through Fairground, the third iteration (v3) of the Vega Protocol testnet. Vega’s restricted mainnet, capped at 13 validators, only allows token staking/delegation and governance voting for the Alpha Mainnet. Fairground allows prospective users to try out some features using testnet assets, including market creation, liquidity provision, and continuous trading through the Vega Console (limited to cash-settled futures through a CLOB).

In the coming weeks, Vega will launch its Alpha Mainnet. Alpha Mainnet market creation and trading will be limited to CLOB-only cash-settled futures, with deposit limits and withdrawal delays for large sums. As the network proves to be secure, reliable and functioning as planned, restrictions will be phased out through on-chain governance.

Competition

Vega is a decentralized exchange (DEX), and given its wide scope, to allow permissionless trading and the creation of virtually any market, it technically competes against almost all DEXs. However, Vega’s focus on margin trading means its core competitors are protocols focusing on perpetuals (e.g., dYdX, Perp), options (e.g., Dopex, Lyra), or a mix of markets such as GMX for spot swaps and perpetuals, and Zeta Markets for options and futures.

Crypto Derivatives Competitive Landscape. Source: Jump Crypto

dYdX is currently far and away the largest on-chain derivatives exchange by trading volume. dYdX only supports one product (perpetuals), has permissioned market creation and an off-chain, centralized order book and trade matching engine. Additionally, the dYdX governance token has virtually no value accrual. Nonetheless, dYdX has a valuable token and does over $1 billion in trading volume every day. Even though a large part of this volume comes from unsustainable trading incentives, some of its success is due to speed and usability that is comparable to centralized exchanges. However, dYdX is moving towards Vega’s model with dYdX V4, which focuses on decentralization (order book, matching engine, market creation, etc.) and distributing trading fees to tokenholders.

From a competitive point of view, dYdX’s transition is mixed news for Vega. While dYdX V4 poses a competitive threat due to the overlap in specifications, it also leaves dYdX vulnerable during its transition. Additionally, we still don’t know if fully decentralized, permissionless, on-chain derivatives really work or can compete against their centralized counterparts, since a lot of dYdX’s volume has been bootstrapped through incentives and centralization. Whether or not Vega, or for that matter dYdX, will be able to attract meaningful volume to (fully) decentralized derivatives exchanges or fall prey to the same problems facing existing protocols, such as user experience (UX) complexity, high latency/flash crashes, and challenges around sourcing/placing liquidity is still an open question.

Risks

Vega tries to solve some of these problems by introducing a modular trading interface (Vega Console), offering 1 second block times with high throughput and creating unique incentives around liquidity. However, there are still some unknowns, including UX friction and security concerns associated with cross-chain bridging and the (required) Vega Wallet. There is also uncertainty around the protocol’s ability to execute high, organic volume across different product types and around the reliability of its first-of-a-kind liquidity model. Additionally, by bootstrapping its own network (blockchain), Vega doesn’t inherit the security or composability/network effects of pre-existing monolithic chains (i.e., Ethereum, Solana, etc.). The security aspect, in particular, can result in a malicious actor taking control over the Vega network by subverting the consensus mechanism. Finally, Vega will face some of the same challenges as other on-chain derivatives exchanges around market manipulation and dubious/unreliable oracle data.

Closing Thoughts

The London Metals Exchange (LME) nickel-trading fiasco could have been averted on an open, decentralized system like Vega.

For starters, on March 7th, the day before the contract cancellations, the price of nickel saw a 30 standard deviation move. Despite the extreme price move, LME governance failed to use its discretion to suspend trading. Vega, on the other hand, uses a transparent, preset threshold to trigger circuit breakers and move the market into a protective auction.

LME’s trades were canceled because the Tsingshan Holding Group, and others, could not meet margin calls and in some cases faced bankruptcy. In Vega, margin is continuously monitored, and stringent risk parameters ensure that the margin posted is always greater than a trader’s loss, guaranteeing solvency. Once a trader’s margin does fall below the minimum collateral requirement (maintenance margin), there is an instant, permissionless liquidation by the protocol, ensuring that the position stops moving against the trader.

Finally, Vega runs on a blockchain. This means orders and liquidations are immutable, and no central authority can easily change the network’s rules.

In the US equity market, options volume is 35 times that of spot, but in crypto, it’s 2%, with most volume coming from centralized exchanges. This number means there is either a lack of demand or a massive opportunity, and given the dearth of good choices, the latter seems more likely. Vega wants to become a leading player in this space by solving the problems that plague on-chain derivatives: fragmentation, high latency, capital inefficiency, and poor liquidity management. If execution matches ambition, Vega has the opportunity to change the face of decentralized derivatives trading and become a centerpiece of DeFi.

This report was commissioned by Vega, a member of Protocol Services. All content was produced independently by the author(s) and does not necessarily reflect the opinions of Messari, Inc. or the organization that requested the report. Paid membership in Protocol Services does not influence editorial decisions or content. Author(s) may hold cryptocurrencies named in this report.

Crypto projects can commission independent research through Protocol Services. For more details or to join the program, contact ps@messari.io.

This report is meant for informational purposes only. It is not meant to serve as investment advice. You should conduct your own research, and consult an independent financial, tax, or legal advisor before making any investment decisions. The past performance of any asset is not indicative of future results. Please see our terms of use for more information.