How is the bid-ask spread formed? The risk pricing logic of gold market makers — Understanding the formation and changes of the spread from the HKEX GDU gold futures

How is the bid-ask spread formed? The risk pricing logic of gold market makers

——Understanding the formation and changes of spreads from HKEX GDU gold futures

Key TakeawaysIn early 2026, the gold market experienced sharp volatility, with the Cboe Gold Volatility Index, which measures expected 30-day gold volatility, rising above 44 at one point, reaching relatively high levels in recent years. When market participants' judgment of the next minute's price movement suddenly becomes unstable, market makers are not simply facing "will gold go up or down," but whether the price will move quickly in an unfavorable direction after each trade. In such an environment, the bid-ask spread is not an arbitrary price difference in the market, but the result of the combined effects of market volatility, trading conditions, and liquidity. This article uses the Hong Kong Exchanges and Clearing Limited (HKEX) USD Gold Futures (GDU) as an example to explain how spreads are formed and how professional liquidity providers maintain stable quotes across different market environments through algorithmic pricing, risk control, and inventory management.

In early 2026, after gold prices hit an all-time high, they quickly retreated, and the expected 30-day volatility reflected in the options market once rose above 44. For ordinary investors, this means gold prices could change significantly in a relatively short period; but for liquidity providers who continuously post bid and ask quotes, it means something more immediate: the price that just traded may become outdated in the next moment.

In such an environment of rapidly rising expected volatility, market-making systems typically need to reassess quote width, order size, and inventory exposure. Changes in the spread are not a temporary judgment about a particular investor, but rather an algorithmic dynamic pricing of market risk.

Market data after the relaunch of GDU shows that under normal trading conditions, liquidity providers can maintain relatively narrow spreads, with the typical bid-ask spread being about 2 to 3 tick sizes, which can further narrow to 1 tick size in calm market conditions.

For GDU liquidity providers including Delta Horizon Capital（德合资本）, maintaining a stable spread is not simply about narrowing the quote distance, but about continuously optimizing the balance between liquidity, risk control, and inventory management. In the process of continuously maintaining two-way quotes, these institutions must constantly answer a seemingly simple but ever-changing question: Given the current market risk, how much distance should be kept between the bid and ask prices?

I. What is the Spread: Transaction Cost and Market Liquidity

A market maker posts both a bid price and an ask price on the order book. The bid price is the price at which the market maker is willing to buy, and the ask price is the price at which the market maker is willing to sell. The difference between the two is the bid-ask spread.

Take the GDU contract as an example: one lot corresponds to 1 kilogram of gold with a purity of not less than 9999. The minimum price fluctuation unit is $0.01 per gram. Assume at a certain moment, the market maker's bid is $130.00 per gram, and the ask is $130.02 per gram, a difference of $0.02 per gram. Since one GDU lot corresponds to 1,000 grams, the notional spread is $20 per lot.

In a market environment with frequent trading, stable and competitive spreads can reduce the immediate cost for investors when trading, and also reflect the adequacy of market liquidity. However, maintaining narrow spreads is not simple—liquidity providers need to continuously manage the impacts of market volatility, adverse selection, and inventory changes.

The market maker does not know whether the next order is from an informed trader with more information or from an ordinary hedging need. If the counterparty has a more accurate judgment on the price direction, the market maker is likely to face an unfavorable price move immediately after the trade—this phenomenon is called "adverse selection." The wider the spread, the greater the compensation the market maker receives for bearing this risk.

Therefore, even if the market is very calm, the spread will not necessarily narrow indefinitely. The minimum price fluctuation unit, transaction costs, hedging costs, and market competition all jointly affect how much quotes can be compressed.

What does this mean for you? The spread between the best bid and best ask that you see on your trading terminal is not a random number, but the result of a market-making algorithm calculating in real-time based on the current market risk level. All else being equal, the narrower the spread, the lower the immediate transaction cost for an investor to cross the bid-ask spread; the actual transaction cost will also be affected by factors such as order size, market depth, and order type.

II. How Volatility Affects Spread Width

Different institutions use models that are not exactly the same, but the common spread adjustment logic in the industry typically considers volatility risk, order flow, and inventory status. So, how does the algorithm judge the current market risk? One of the most direct factors affecting quote width is market volatility.

The system may combine information such as recent realized volatility, implied volatility from the options market, price jumps, and event risk to assess the likelihood of a significant price change over a short future period. The higher the volatility, the higher the uncertainty that prices will experience large changes in the near term.

When the market environment is stable, price changes are relatively predictable, and liquidity providers can usually maintain tighter quotes, allowing investors to complete trades at a lower immediate cost. In periods of sharp volatility, dynamically adjusting quotes helps the market maintain trading continuity.

Therefore, in a calm market with low volatility, institutions may strive for transactions through more competitive spreads; when risk rises, they need to increase the risk premium in their quotes. Different products' liquidity provider programs may have their own quote obligation arrangements, and the specific quote requirements applicable to GDU should be subject to the relevant liquidity program and participation arrangements.

This logic of adjusting spreads based on risk is particularly important for contracts like GDU that are linked to global gold prices. International gold prices often experience instantaneous volatility during major data releases, geopolitical events, or central bank policy statements. The market-making algorithm needs to complete volatility assessment and spread adjustment within an extremely short time.

What does this mean for you? The spread cost you pay for the same trade at different times may be different. This is not because the market maker is "targeting you," but because the algorithm adjusts the risk pricing in real-time based on the market volatility at that time. All else being equal, the calmer the market, the more conditions institutions usually have to offer more competitive spreads.

III. Order Flow Characteristics: Not All Trades Are "Equal"

But volatility can only answer "how dangerous the market is overall," but cannot answer another key question: whether this particular trade itself may contain information advantage. Another important input for the spread algorithm is order flow characteristics—assessing the adverse selection risk of the current trade based on statistical signals such as short-term price changes after a trade, order book imbalance, and order arrival patterns.

Take a simplified example: if a market maker finds that every time it sells, gold prices immediately rise, and every time it buys, gold prices immediately fall—this indicates that it is on the unfavorable side after each trade. This pattern suggests that the counterparties as a whole may have more accurate market direction judgments. To address this risk, the market-making system may proactively widen the spread or adjust the quote size.

An actual system does not draw immediate conclusions from a single trade, but continuously tracks the trade direction, subsequent price changes, order book imbalance, and order arrival characteristics over a period. When multiple signals collectively indicate rising adverse selection risk, institutions may widen the spread, reduce quote quantity, or adjust quote positions.

It should be noted that the algorithm judges the statistical relationship between order flow and subsequent price changes over a period, rather than identifying or evaluating the trading purpose of a specific investor. Regardless of the intention behind the orders, if orders in the same direction appear concentrated over a period, accompanied by subsequent unfavorable price moves, it may increase the system's estimate of overall adverse selection risk.

What does this mean for you? The market maker's spread adjustment is not targeted at individual investors, but is a statistical judgment made by the system based on post-trade price trends and order flow characteristics. Changes in the spread reflect a dynamic assessment of the overall trading environment, not a temporary reaction to a specific order.

IV. How Inventory Status Affects Quotes

In addition to risks from the market and counterparties, market makers also need to manage the risk arising from changes in their own positions. Volatility and adverse selection risks mainly affect how much risk compensation a market maker needs, while inventory status more often affects the center of quotes and the relative position of the bid and ask sides.

When the short position deviates from the target level, the market maker may shift the bid and ask prices upward as a whole—this aims to reduce the probability of further sell trades while increasing the probability of buy trades, gradually reducing the existing short position. Conversely, when the long position deviates, the prices are shifted downward as a whole. Whether the overall spread also widens depends on the combined effects of factors such as volatility, liquidity, and risk limits.

Therefore, the bid and ask quotes that investors see on the order book are actually the result of a combined calculation of three factors: volatility risk, adverse selection risk, and inventory status. When volatility, adverse selection signals, and inventory pressure are all low, institutions are typically more capable of narrowing the spread; when one or more of these risks rise, the spread may widen, and the quote quantity and position may also be adjusted accordingly.

For HKEX gold futures liquidity providers, mature spread management requires a comprehensive assessment of volatility risk, adverse selection risk, and inventory risk, with dynamic quote adjustments based on market changes. These three dimensions often deteriorate simultaneously during market turbulence, creating compound pressure on the algorithm.

What does this mean for you? The spread is never the output of a simple formula; it is the superposition of volatility risk, adverse selection risk, and inventory risk. Understanding these three layers of logic helps you understand why the spread is sometimes wide and sometimes narrow—not because the market maker is adjusting arbitrarily, but because each layer of risk is changing in real time.

Conclusion: The Spread is a "Thermometer" of Market Risk

The sharp volatility in the gold market in early 2026 reminds all market participants: the spread is not set arbitrarily by market makers. When the gold volatility index spikes, the market-making algorithm adjusts quote width, which is typically a dynamic response to rising volatility, adverse trades, and inventory risk. Without dynamic risk management capabilities, it may be difficult for market liquidity to remain stable in extreme environments.

For HKEX gold futures liquidity providers, the spread algorithm is a crucial link connecting market risk with the institution's ability to continuously quote. The value of professional liquidity providers is not merely providing two-way quotes, but using technology systems and risk management capabilities to keep markets trading continuously across different market environments and help investors achieve more efficient price discovery.

For investors, understanding the spread algorithm is not about predicting the direction of spreads, but about understanding a basic fact: every bid-ask spread you see in the market is not a static label, but a joint pricing of volatility risk, information asymmetry risk, and inventory risk at that instant. Professional market-making institutions, by continuously managing these risks, provide more stable liquidity and more efficient price discovery to the market, thereby helping reduce investors' transaction costs in normal market environments. Futures and derivatives carry leverage risk, and technical discussions of spread models do not constitute a guarantee of future market performance or investor profit or loss.

References:
[1] Reuters Breakingviews, "Options muddy gold's value as gauge of global risk", February 2026.
https://www.reuters.com/commentary/breakingviews/options-muddy-golds-value-gauge-global-risk-2026-02-02/

[2] Hong Kong Exchanges, USD Gold Futures Contract Specifications and Liquidity Provider List.
https://www.hkex.com.hk/Products/Listed-Derivatives/Commodities/USD-Gold?sc_lang=en

[3] Hong Kong Exchanges, Market Maker / Liquidity Provider Obligations and Incentives.
https://www.hkex.com.hk/Products/Listed-Derivatives/Market-Maker-Program/Market-Maker-Obligations-and-Incentives?sc_lang=en

[4] Xinhua Finance, "HKEX USD Gold Futures Optimization and Upgrade: First Day Volume Sets Historical Record", July 6, 2026.

Disclaimer: This material is provided by Delta Horizon Capital（德合资本）for institutional investor reference only. It does not constitute any express or implied investment advice, offer, or solicitation. The data and information contained in this report are sourced from public market channels. Delta Horizon Capital（德合资本）makes no warranty as to their accuracy or completeness. Past performance is not indicative of future results. Market risk exists, and investment should be made with caution.

Delta Horizon Capital（德合资本）is a quantitative trading firm headquartered in Hong Kong, with its technology and research team based in North America, focusing on quantitative trading, two-way quoting, and market liquidity provision. The company has long been deeply involved in diversified markets including precious metals, stock index futures, foreign exchange and interest rates, energy, and commodities. Relying on mature quantitative pricing models, a rigorous risk management system, and low-latency trading infrastructure, it provides stable, efficient, and flexible liquidity services to institutional clients. The core team members collectively have over 30 years of experience in international futures, options, and derivatives markets. Since 2017, the Dehe trading team has successively obtained the first batch of on-exchange market maker qualifications for base metals, gold futures, A50 stock index futures, and RMB currency futures from the Hong Kong Stock Exchange, and has received multiple market maker and liquidity provider awards from HKEX. Delta Horizon Capital (Hong Kong) Limited is currently listed on the HKEX USD Gold Futures (GDU) liquidity provider list. For more details, please visit the official website: www.deltahorizoncapital.com