Figures are in million troy ounces (Moz) unless stated. Supply and demand numbers come from the World Silver Survey 2026, published in April by the Silver Institute, an industry association whose members include silver miners and refiners, with data compiled by the independent consultancy Metals Focus. Profoundd read the sources on Nov. 6, 2026. This is commentary, not investment advice; check figures against the latest releases before relying on them.
Mines alone do not cover demand
Mines produced 846.6 Moz of silver in 2025, while total demand was 1,130.6 Moz, according to the survey. Mine output covered 74.9% of demand, leaving a gap of 284.0 Moz. Most of that gap is filled from other sources: recycling supplied 197.6 Moz, the most in 13 years, and producer hedging added 44.7 Moz, bringing total supply to 1,090.4 Moz. Even with those, demand exceeded supply by 40.3 Moz, or 3.6% of demand, the fifth straight annual deficit. The shortfall was far smaller than in earlier years: 254.0 Moz in 2022, 200.1 Moz in 2023 and 137.9 Moz in 2024. The survey forecasts a sixth deficit of 46.3 Moz in 2026, which would bring the cumulative shortfall since 2021 to 762.1 Moz, met by drawing down stocks above ground.
That forecast is already being questioned by the firm that produced it. In an October interview, Metals Focus director Matthew Piggott put the 2026 deficit at 35.4 Moz, and for 2027 the firm forecast what GOLDINVEST described as a “theoretical market surplus of 37.3 million ounces,” GOLDINVEST reported, because solar makers’ savings in silver per panel now exceed the growth in installations. The Silver Institute and Metals Focus usually publish an interim update in November; until then the April survey remains the published forecast.
Has production been falling?
Partly. Mine output peaked at 900.1 Moz in 2016, according to the survey’s 2025 edition, and the 2025 total is 53.5 Moz (5.9%) below that peak. The low point in the current data was 790.3 Moz in 2020. Output has risen since, by 1.6% in 2024 and 3% in 2025, and Metals Focus expects 2026 to be essentially flat, at 844.1 Moz (down 0.3%). Annual figures in the 2026 edition revise earlier ones slightly; 2024, for example, is 823.6 Moz, against 819.7 Moz in the 2025 edition.
The survey points to structural reasons supply responds slowly to price. Only 26% of mined silver now comes from primary silver mines, a record low; the rest is a by-product of lead and zinc mines (249.1 Moz), copper mines (237.3 Moz) and gold mines (134.7 Moz). Output from primary silver mines fell for a third straight year in 2025, “following the closure of some mines due to reserve depletion and lower output at numerous, sizable operations.” Mexico, the largest producer, fell 5% to 172.9 Moz, which the survey attributes to “operational disruptions, policy shifts and declining grades.” Against that, record copper and gold prices are encouraging the base-metal and gold miners that produce most silver to expand, and the 2025 increase came largely from copper mines in Peru and the ramp-up of Polymetal’s Prognoz mine in Russia.
Where the silver goes
Industry used 657.4 Moz in 2025, 58% of demand, down 3% in its first decline since the pandemic. Within that, electrical and electronics took 449.5 Moz, of which solar panels used 186.6 Moz; brazing alloys and solders took 50.5 Moz; and other industrial uses, led by catalysts for ethylene oxide, took 157.4 Moz. Jewelry used 189.3 Moz (down 8%), silverware 42.1 Moz (down 21%) and photography 24.2 Moz. Coins and bars took 217.7 Moz, up 14%. Separately, exchange-traded products backed by physical silver added 278.1 Moz; the survey reports those flows beside the demand total rather than inside it, and counting them the 2025 gap widens to 318.4 Moz.
Solar has been the growth engine and is now the main drag. Silver use in solar panels more than doubled from 82.8 Moz in 2020 to 197.5 Moz in 2024, then fell 6% in 2025 and is forecast to fall 19% to 151.0 Moz in 2026. The survey says rising silver prices pushed silver from 8–10% of the cost of a solar cell at the start of 2025 to more than 20%, and manufacturers responded by cutting silver: pastes with only 10% silver content entered high-volume production in late 2025, and the survey expects mainstream cells to use below 5 milligrams of silver per watt by 2027. The industry’s own International Technology Roadmap for Photovoltaic (ITRPV), published in June, found a median of about 10 mg/W for the most common cell type in 2025 and expects 6.3 mg/W within ten years. Its summary: “Silver reduction works!” Growth elsewhere is real but smaller: the survey credits AI data centers, cars and power-grid spending with supporting electronics demand, and Oxford Economics, in a December 2025 report produced for the Silver Institute, forecasts automotive silver demand to grow 3.4% a year from 2025 to 2031, to about 94 Moz.
The next ten years
Profoundd found no published forecast of total silver demand out to 2035. The Silver Institute survey forecasts one year ahead and describes the longer term in words: price-led thrifting and lower jewelry and silverware buying “should see consecutive annual net reductions in demand,” while mine production “should continue to grow slightly,” so the deficit “should erode over time” toward “a more balanced market state.” In the same passage it says a swing to a sustained surplus would take “a rather extreme combination of changes,” and that data centers, infrastructure, cars and aerospace could slow the erosion. Metals Focus expects industrial demand to fall from 640.2 Moz in 2026 to 618.2 Moz in 2027, GOLDINVEST reported.
To answer the question of how much demand could grow, Profoundd built a scenario from those documented components. It is Profoundd’s arithmetic, not an official forecast. Solar is the swing factor. Installations stalled in 2025 (the ITRPV counted 706 gigawatts of modules shipped, against 703 in 2024), and BloombergNEF expects 584 GW of installations in 2026, 10.8% below 2025, as China falls, pv magazine reported. The International Energy Agency’s main case expects solar to make up nearly 80% of about 4,600 GW of renewable capacity added from 2025 to 2030. In the scenario, solar silver is 2025 use multiplied by an installation factor and a silver-per-watt factor: the low case holds installations at BloombergNEF’s 2026 level and halves silver per watt, in line with the survey’s 2027 target; the middle case holds installations at 2025 levels and cuts silver per watt by the 37% the ITRPV expects; the high case assumes installations 20% above 2025 and thrifting at half the ITRPV pace. Other categories follow their 2017–2025 trends or ranges, as set out below.
| Moz | 2025 actual | 2035 low | 2035 middle | 2035 high | Assumption (low / middle / high) |
|---|---|---|---|---|---|
| Solar panels | 186.6 | 83.2 | 117.6 | 182.5 | installations x silver per watt, as described above |
| Other electrical & electronics | 262.9 | 262.9 | 294.9 | 367.3 | flat / 2017–2025 trend, 1.2% a year / 3.4% a year (Oxford Economics auto rate) |
| Brazing alloys & solders | 50.5 | 50.5 | 50.5 | 50.5 | flat, as since 2017 |
| Other industrial | 157.4 | 157.4 | 185.5 | 185.5 | flat / 2017–2025 trend, 1.7% a year |
| Photography | 24.2 | 16.8 | 16.8 | 16.8 | 2017–2025 decline continues, 3.6% a year |
| Jewelry & silverware | 231.4 | 192.9 | 231.4 | 249.1 | 2026 forecast held / 2025 held / 2017–2025 average |
| Coins & bars | 217.7 | 155.5 | 221.8 | 339.5 | 2017–2025 low / average / 2022 peak |
| Total demand | 1,130.6 | 919.2 | 1,118.5 | 1,391.2 | change from 2025: −211.4 Moz (−18.7%) / −12.1 Moz (−1.1%) / +260.6 Moz (+23.0%) |
Rows may not add exactly because of rounding. 2025 figures: World Silver Survey 2026. Scenario: Profoundd, Nov. 6, 2026.
On those assumptions, total demand in 2035 lands between 919.2 Moz and 1,391.2 Moz, and the middle case is roughly flat, about 1.1% below 2025. The high case, 23.0% above 2025 or about 2.1% a year, requires solar thrifting to slow sharply and coin and bar buying to return to its 2022 record; the low case requires both to go the other way. Measured against supply, if mine output and recycling stayed at the 2026 forecast of 1,055.4 Moz combined, the middle case would leave a shortfall of 63.1 Moz a year, 5.6% of demand, to be met from stocks, more recycling or new mines; the low case would leave a surplus of 136.2 Moz and the high case a shortfall of 335.8 Moz. Supply is not fixed either: higher prices pulled recycling to a 13-year high in 2025, and the survey expects it to rise 7% to 211.3 Moz in 2026.
Profoundd’s read
The first half of the premise holds: mines have not produced enough silver to meet demand in any year in the survey’s data, and the market has run five straight deficits that used up stocks. Mine output is below its 2016 peak, but it has risen for two years, and the shrinking deficit since 2022 came mainly from falling demand: from 2022 to 2025 demand fell 153.5 Moz (12.0%), while total supply rose 60.3 Moz (5.9%). Whether demand grows over the next decade turns mostly on solar: cheaper, thinner use of silver per panel has so far outpaced the growth in panels, and the people who compile the industry’s figures forecast total demand to fall again in 2026 and industrial demand to fall in 2027. Sustained growth from here would need solar installations to resume growing faster than thrifting, or investment buying to stay near records; the scenario shows how wide that range is.
