When a gap emerges between Shanghai and international gold prices, the first step is to confirm you are comparing the same type of price. Shanghai quotes denominated in RMB per gram cannot be directly subtracted from international quotes in US dollars per troy ounce. Nor can the closing price of one market be contrasted against the latest quote from another market recorded hours later as proof of a sudden surge in local demand. Only when currency, unit, timestamp and product specifications are aligned can the remaining spread warrant further analysis.
First clarify what “Shanghai gold price” and “international gold price” each refer to
The Shanghai Gold Benchmark Price, spot outright transaction prices on the Shanghai Gold Exchange (SGE), and gold futures prices on the Shanghai Futures Exchange (SHFE) may all appear in news reports, yet they are distinct indicators. The SGE Au99.99 product page specifies that trading is spot outright, quoted in RMB per gram. The parameter page for the Shanghai Gold Benchmark Price states that SHAU adopts a centralised pricing mechanism. Even when both are displayed in RMB per gram, their price formation timelines and mechanisms are not identical.
“International gold price” may refer to spot quotes on the London market, composite prices from a data provider, or COMEX futures for a specific contract month. Futures prices incorporate factors such as time to expiry, while over-the-counter quotes are split into bid and ask sides. When recording prices, always save the product name and data source alongside the figure; avoid labelling them merely as “domestic” and “international”.
Currency conversion must be paired with weight unit conversion
International gold is commonly quoted in US dollars per troy ounce. For valid comparison, first convert troy ounces into grams to get the US dollar-per-gram equivalent, then multiply by the USD/CNY exchange rate captured at the same moment. The ounce here is the troy ounce used for precious metals, not the avoirdupois ounce used for general commodities. If the exchange rate is quoted as how many RMB one US dollar buys, a rising value means one US dollar can be exchanged for more RMB.
After conversion, you may calculate the absolute spread as “Shanghai quote minus the RMB-per-gram equivalent of the international quote”, or divide the spread by the converted international price to get a percentage difference. The two metrics answer separate questions: when the overall gold price rises, the same percentage spread can translate to a larger per-gram gap. Therefore, when discussing whether the spread has widened, specify whether you are referencing absolute value or percentage. Do not switch arbitrarily between the two standards.
Quote timestamps and bid/ask sides alter the observed gap
Centralised benchmark results reflect a specific pricing session, while real-time continuous quotes fluctuate constantly. If international gold moves sharply after the Shanghai pricing window closes, subtracting the fixed benchmark result from the latest international quote introduces a time lag bias. A more meaningful comparison samples prices within the same time window, and notes whether the data is delayed market data, daily closing price or settlement price.
The bid and ask sides must also match. Suppose after conversion to the same unit, the mid-price in Market A is 600 RMB and the mid-price in Market B is 602 RMB, showing an apparent 2 RMB gap. If buying in Market A actually costs 601 RMB and selling in Market B only nets 601 RMB, the mid-price spread shown on screen offers no executable arbitrage room. This example only illustrates quote mechanics and is not real market data. Executable volume and market depth may further change the outcome.
Differences in delivery and costs mean spreads cannot be fully attributed to demand
Even when both markets trade gold, they may differ in ingot specifications, purity, delivery location, settlement timeline and eligible participants. Converting physically deliverable gold from one region into a form acceptable in another may involve transportation, insurance, warehousing, assay or refining procedures. Tax treatment must be verified against the specific product and business rules. Retail jewellery prices including fabrication premiums should never be directly compared against wholesale ingot prices.
These conditions determine whether a price divergence can quickly converge via trading. Seeing two separate quotes does not mean a single trader can execute trades in both markets simultaneously, allocate gold stock and complete cross-border fund settlement. Capital tenure, foreign exchange conversion costs and cross-border compliance rules also restrict feasible arbitrage operations. For this reason, a recorded premium is not equivalent to accessible arbitrage profit, as a set of practical execution requirements stands between them.
How to judge whether the remaining spread is analytically meaningful
To track spread changes over time, stick to one fixed Shanghai product, one fixed international data source and a consistent time window. Keep the exchange rate conversion, weight unit and bid/ask handling standardised. Only then can you tell whether spread movements stem from genuine market shifts or inconsistent sampling methods. If only asynchronous data is available, describe the performance of each price separately, rather than presenting their difference as an accurate real-time premium.
Once a consistent-calibre spread shows sustained movement, cross-reference physical trading volumes, inventory levels or import statistics for local clues. A single day’s spread expansion can raise questions about demand or supply, but cannot confirm the root cause on its own. Lucky Gold’s guide distinguishing London Gold from COMEX gold helps you identify the nature of international quotes. Cross-market comparison additionally requires aligned timestamps, exchange rates and delivery terms before figures become interpretable.