Loading...
Loading...
Gold and silver premiums do not move randomly across the calendar. Cultural celebrations, harvest cycles, and institutional portfolio rhythms create repeating seasonal patterns in how much buyers in Shanghai, Mumbai, and London pay above the COMEX benchmark. This page visualizes those patterns from historical data so you can see which months consistently carry higher or lower premiums in each market.
Each bar represents the average premium (or discount) for that market in that month, expressed as a percentage versus the COMEX futures benchmark. A bar above zero means the local market has historically traded above COMEX; a bar below zero means a discount. Colors identify markets: red for SGE (Shanghai), orange for MCX (Mumbai), blue for LBMA (London). Taller bars indicate stronger seasonal demand pressure. The sample size shown in the tooltip matters: months with many data points (typically shown as "High confidence") are statistically more reliable than months with sparse observations. To compare metals, use the toggle at the top of the chart.
The Shanghai Gold Exchange is the world's largest physical gold market, and its premium profile is shaped above all else by the Chinese cultural calendar. The Lunar New Year (falling in late January or February) drives the single largest seasonal spike: retailers pre-stock gold jewellery and gift bars for weeks in advance, and private buyers purchase gold as a traditional wealth-preservation gesture for the new year. Premiums on the SGE typically start rising in November and peak in the two weeks before the holiday. A secondary pulse arrives in September and October around the Mid-Autumn Festival, when gold mooncake boxes and gift sets are in high demand. Wedding season also contributes — in China, gold jewellery is a near-universal wedding gift, and the peaks in spring (April-May) and autumn (September-October) correspond to popular wedding periods. Taken together, these cultural drivers mean that the months of January, February, September, and October tend to carry the highest SGE premiums in a typical year.
India is the world's second-largest gold consumer, and MCX premiums are heavily influenced by an unusually dense festival calendar. Akshaya Tritiya — the spring harvest festival considered the most auspicious day of the year to buy gold — typically falls in April or May and produces a sharp demand spike. Diwali, the festival of lights in October or November, is the second most important gold-buying occasion: families traditionally purchase gold coins, small bars, and jewellery as Lakshmi offerings. The wedding season in India runs from November through February, with a secondary wave in April and May; gold jewellery is a core component of the dowry and gifting culture, making these months structurally bullish for MCX demand. A quieter period occurs in July and August during Shraadh (a mourning period when auspicious purchases are deferred), creating one of the few reliable seasonal dips in Indian demand. Understanding these festivals is essential for reading the MCX premium chart: a spike in April-May almost always reflects Akshaya Tritiya buying, while an October-November spike points to Diwali and pre-wedding stock-building.
The London Bullion Market Association operates through global bank and dealer networks, and its premium movements are less driven by cultural festivals and more by institutional trading rhythms and physical arbitrage economics. The most consistent seasonal pattern for LBMA is a quiet summer — June, July, and August — when European and North American trading desks are thinly staffed due to summer holidays, volumes contract, and the bid-offer spreads on loco London metal can widen or narrow in ways that affect the computed premium. Lower liquidity can temporarily push the LBMA premium modestly above or below what fundamental economics would suggest, simply because fewer market-makers are present to absorb imbalances. The end of the Northern Hemisphere summer (September) typically sees a pick-up as institutional demand restarts, particularly from central bank programmes and sovereign wealth funds that run regular purchase schedules. While the LBMA premium swings tend to be smaller in absolute magnitude than SGE or MCX, the summer-to-autumn shift is one of the more reproducible patterns across multiple years of data.
December and January carry a distinct dynamic that is more macro-financial than cultural: institutional year-end and year-start portfolio rebalancing. Fund managers closing their books in late November and December often reduce commodity positions to lock in annual performance, which can soften premiums by releasing supply to the market. Meanwhile, the January effect in gold — fresh allocation mandates at the start of a new fiscal year — can cause a brief premium uptick as buyers re-enter simultaneously across SGE, MCX, and LBMA. Central bank purchases, which follow budget cycles and end-of-year compliance windows, add another layer: many sovereign buyers run purchases in the final weeks of the fiscal year to hit allocation targets, then pull back in January and February. For the LBMA market, the Commitment of Traders data from related futures markets shows consistent positioning changes in late November through early January, reinforcing the notion that the December-January window is as much about financial mechanics as it is about seasonal demand.
Seasonal patterns are tendencies, not guarantees. A year dominated by a major geopolitical event — sanctions, capital controls, currency crises, or a pandemic — can completely override the cultural demand cycle. The longer the historical period you select, the more observations contribute to each monthly average and the more stable the pattern becomes, but even a three-year window is short by the standards of macroeconomic time series. Recent data always carries more weight in real-world decision-making than a multi-year average. Additionally, the premium values here are derived from exchange prices sampled at specific intervals, not continuous intraday data, so very short-lived spikes may not be captured.
This page is for informational purposes only. See our Disclaimer for more information.