The gold–silver ratio is one market price divided by another. At a stated time, take the gold print in dollars per troy ounce and divide by the silver print in dollars per troy ounce. The result is how many ounces of silver equal one ounce of gold at those two prints. It is a dated snapshot. It is not a fair-value claim.
What the ratio measures
Call gold G and silver S, both in the same currency per troy ounce. The ratio is G ÷ S. If gold is $2,000 and silver is $25, the ratio is 80. Eighty ounces of silver then have the same dollar value as one ounce of gold at that pair of prints. Change either print and the ratio changes. There is no third hidden input.
The two prices must share a clock. A London gold PM fix against a COMEX silver nearby from another session is a mixed snapshot. This site’s desk shows a live ratio from the same live prices. This page uses named prints and year averages so the arithmetic can be checked. This site’s published year-average price series is an LBMA/COMEX annual average for each metal; dividing those two averages gives a year-average ratio, which is not the same as any single day’s print.
Historically, states also wrote a mint ratio: a legal number of silver units per gold unit. The Coinage Act of 1792 used 15 to 1. Later statutes used 16 to 1. That legal ratio is a mint rule. It is not the market ratio. When the two diverge, the legally overvalued metal tends to stay in coin and the other tends to leave — the ordinary bimetallic problem, told as narrative under bimetallism. This page keeps the market quotient.
What the ratio does not measure
The ratio does not measure a natural law. Geology, mine supply, industrial use, and monetary demand all affect the two prices. None of them is “the” ratio. A number near 15 in a mint statute, or near 17 on a January 1980 tape, does not bind a later tape.
The ratio does not measure whether silver is behind or gold is ahead. It does not contain a mean that prices owe a return to. It does not forecast a catch-up. Those sentences are the ones this page will not make. A high ratio means gold’s dollar print is large relative to silver’s dollar print at that date. A low ratio means the opposite. That is the whole claim.
The ratio also does not measure industrial tightness by itself. Silver has a large fabrication use; gold’s fabrication use is smaller relative to its monetary stock. A squeeze in one metal can move the quotient without saying anything about the other metal’s “true” value. Keep the definition narrow so the snapshot stays honest.
1980 and 2011 as anchors
Two well-documented peaks are the anchors on this page. They are dates, not destinies.
In January 1980 the London gold PM fix printed $850 on 21 January. Silver’s nearby extreme in that same month, used on this site’s 1980 desk print, is $49.45 on 18 January. Those two named prints give $850 ÷ $49.45 = 17.2. That is a peak-week snapshot during the Hunt-era silver run, whose break is told as Silver Thursday. The 1980 year averages on this site are $612.56 gold and $20.98 silver → 29.2. The January tape and the year average are different snapshots. Both are arithmetic.
In 2011 the London gold PM fix printed $1,895 on 6 September. Silver’s 2011 year average on this site is $35.12; gold’s is $1,571.52 → a year-average ratio of 44.7. April 2011 saw silver nearby prints in the high forties while gold was still below the September fix, so intra-year ratios ran from the mid-thirties into the forties depending on the day. The useful habit is to name the two prices and the date. “The 2011 ratio” without a clock is a blur.
Year-average snapshots from the same series
Using only this site’s published year-average prices — so both metals share a method — a few more dated quotients sit in one table. Again: arithmetic, not a path.
- 1971: $40.62 ÷ $1.39 = 29.2.
- 1980: $612.56 ÷ $20.98 = 29.2 (year average; January peak prints ≈ 17.2, above).
- 2000: $279.11 ÷ $4.95 = 56.4.
- 2011: $1,571.52 ÷ $35.12 = 44.7.
- 2020: $1,769.64 ÷ $20.55 = 86.1.
- 2024: $2,386 ÷ $28.27 = 84.4.
Those six rows show that a year-average ratio can sit near 29 in two different decades and near 84–86 in two later years without that fact implying a return trip. The 1980 January print near 17 is a third kind of snapshot: a peak-week pair, not a year. Keep the labels on the numbers.
How to read a snapshot
Name the two prices, the venue or series, and the date. Say whether you used a fix, a nearby future, or a year average. Then divide. If you compare 1980 to 2011, say which 1980 and which 2011. The markets hub orients this fact page. Official gold’s leftover U.S. book rate is a different fact, on official gold book value. Central-bank tonnes are a third fact. Identified bar-and-coin offtake by country is a fourth, on physical silver demand by country.
History’s job is the older legal ratio and the 1980 squeeze narrative. This page’s job is the market quotient. Practice’s job is bars, coins, and premiums. Mixing the three produces slogans. Keeping them apart produces a number you can check.
Nothing here is a reason to prefer one metal. Nothing here is a mean, a band, or a catch-up clock. The gold–silver ratio measures a dated pair of prints. That is the claim, and that is the stop.
If you want the older legal ratio — bimetallism, both metals legal at a fixed mint ratio — as a statute fight, open bimetallism. If you want the 1980 squeeze as a narrative, open Silver Thursday. Those are history episodes. They explain how a mint number or a concentrated position entered the record. They do not rewrite the definition on this page. A mint ratio is a law. A market ratio is a quotient. Keep the two labeled when you move between History and Markets.