Rome’s Incredible Shrinking Coin

On April 11, 2010, a metal detector in a field near Frome, England, gave what its owner called a “funny signal.” Fourteen inches down sat a clay pot holding 52,503 Roman coins, about 350 pounds of them, roughly two grown men. Nearly every one was bronze pretending to be silver, like costume jewelry at a wedding: shiny in the photos, and turning everyone’s fingers green by the end of the reception.

That pot is Roman currency debasement, buried. Between Augustus and the year AD 300, the denarius, Rome’s silver coin, slid from roughly 98 percent silver to about 5, without once changing its name or its face value, and always with an emperor’s head on the front. The coin kept its size. The silver left a little at a time, like a kid’s lemonade stand that keeps the price at a quarter and swaps one lemon a summer for tap water, until it’s selling a cup of cold water with a very confident sign.

The pot took archaeologists three days and 66 bags to lift. In its day it held about four years of a Roman soldier’s pay, which sounds like a fortune until you check what the coins were made of.

Did Rome invent shrinkflation?

Not quite. Rome did something bolder: it shrank the money. You already understand how that works, because you have stood in a supermarket holding a bag of chips that costs what it always cost and weighs a little less than it used to. Same bag. Same price. Mostly air, and nobody wrote to tell you.

Shrinkflation runs on a blind spot: we read the price on the front and ignore the weight on the back, and nobody marches on the capital over three fewer chips. Rome aimed the same blind spot at the money itself. Picture your twenty-dollar bill still green, still the same size, still saying twenty, and buying fifteen dollars of groceries. The bill doesn’t mention it either.

The rules were short enough to fit on a coin. Keep the name, the face value and the emperor’s head. Change only the part nobody could check without a furnace.

How much silver was in a Roman denarius?

Here is the slide, in the figures coin graders and historians use (approximate, with sources at the bottom). Every line is a coin that looked like the one before it.

• Augustus: about 3.9 grams, 97.5–98% silver. That’s about $6.64 of silver at the October 24, 2026 price, roughly what a movie theater charges for a small popcorn, hold the butter. For practical purposes it was a lump of silver with a portrait on it, trusted from the fog of Britain to the deserts of Syria. A legionary was paid in these. So was everybody else.
• Nero, AD 64: about 3.4 grams, 93.5% silver. The first serious trim, and the template for every one after it: small, permanent, and announced to nobody, like a streaming service raising its price a dollar and hoping you never open the email.
• Trajan: roughly 89–90% silver. Still respectable, the way a sweater with one loose thread is still a sweater.
• Septimius Severus: down to about 57% during his reign. Half silver, half hope.
• Caracalla, AD 215: a new double-denarius coin, the antoninianus, at 52% silver. Hold that thought.
• Gallienus: a copper coin with a thin silver wash, at 5% or less and in some cases about 2.5%. The silver was now mostly there for morale.
• Aurelian, early 270s: coins stamped XXI, which most likely means twenty parts copper to one part silver, or about 5 percent (the meaning is still debated). The coin had started printing its own ingredients label.
• Around AD 300: the denarius itself at about 5% silver. A silver coin in the sense that a decaf is a coffee.

Read down the list and notice what never appears in it: a line that says and then Rome announced this. The new coins arrived dressed as good news, and the rest happened the way a waistline grows: one Thanksgiving at a time.

In kitchen terms: pour a gallon of milk and let it go the way of the denarius. By AD 300 the jug holds about six and a half ounces of milk, a little less than a cup, and the rest is water. The family cat has filed a complaint.

Picture the announcement Rome never made: “The Imperial Mint is proud to unveil the new denarius, featuring a fresh portrait of the emperor, a laurel wreath and, in a first for Roman money, almost no silver. Please do not bite it.”

What was the antoninianus, and why was it a pay cut?

In AD 215 Caracalla tried the move every debasing government eventually finds. Instead of shaving the old coin again, he issued a bigger one. The antoninianus was declared to be worth two denarii. It weighed about 1.5 times as much as one, and it held less than 70 percent of the silver in two. A coin that called itself a double was really a coin and a half with ambitions, like a hotel “double room” that turns out to be one bed and a very optimistic fold-out couch.

Picture payday on the frontier. Lamplight in a leather tent, the smell of wet wool and woodsmoke, a soldier who has been paid in denarii his whole career. He’s handed a shiny new coin, bigger, with the emperor wearing a radiate crown, a spiky sunburst the old coins never had. Two denarii, he’s told. It looks like a raise. It feels like a raise. It’s a pay cut of around 30 percent with a bow on it, and it works for exactly as long as it takes somebody in the camp to find a scale.

Into the weeds: the arithmetic of a fake double

Two old denarii weigh two units; the new coin weighed about 1.5. At the same purity that alone would be a quarter short. The antoninianus was also less pure: at launch it weighed about 5.1 grams at 52% silver. NGC puts its silver at under 70 percent of two denarii, which is where our “around 30 percent” comes from. That last step is our own arithmetic, and the figures vary a little between sources and between individual coins, which were struck by hand.

By Gallienus the antoninianus had become a copper coin with a silver wash, and the wash wore off with handling. The currency confessed in your pocket. A shiny coin went dull, then pinkish, then brown, as the copper showed through like a bald spot through a comb-over. Imagine a twenty that fades a little every time you buy gas, until one morning you’re at the pump holding a blank green rectangle and the cashier is holding an opinion.

The pot in Frome is full of them: coins struck between AD 253 and 305, the comb-over years.

What happened when Rome fixed prices?

Prices rose, which is the one thing in this story nobody should find surprising. In AD 301 the emperor Diocletian answered with the Edict on Maximum Prices, setting legal ceilings on well over a thousand products and services, with the death penalty for profiteers.

The edict survives in fragments of carved stone, and it reads like a menu written by a very angry accountant. In Antony Kropff’s 2016 translation, the most anyone may charge for a pound of pork is 12 denarii. A haircut is capped at 2, and a farmhand’s day of work at 25, meals included. There is also a maximum price for a lion: 150,000 denarii. A farmhand would need 6,000 days of wages, about sixteen years without a weekend, to buy one at the ceiling.

This is real. Somebody in the imperial civil service sat down and priced a lion. Presumably he went home that night, and when his wife asked how work went, he said one word: lions.

The preamble explains who was to blame, and it wasn’t the mint. “Greed raves and burns and sets no limit on itself,” it says. Follow the chain. The state had spent two centuries taking the silver out of the coin, and merchants, finding that each coin bought less, asked for more of them. The state studied this carefully and located the culprit: the merchants. Asking became a capital offense. It’s a teenager who waters down the family orange juice, then grounds Dad for pouring a bigger glass.

(“By order of the emperor: a pound of pork now costs 12 denarii. The pig has been informed.”)

The pig was not persuaded, and neither was anyone selling one. Lactantius was a Christian writer with his own reasons to dislike Diocletian, so read him with that in mind. His report still rings true: “much blood was shed for the veriest trifles; men were afraid to expose anything to sale, and the scarcity became more excessive and grievous than ever.” Within a few years the edict was mostly ignored.

The number on the sign changes, and the goods behind the counter refuse to follow it. Price controls do this reliably. A government that has debased its money will often decide that inflation is a behavior problem among the people who noticed.

As for the man who priced the lion, Diocletian did something no emperor before him had done: he quit. He abdicated on May 1, 305, and retired to his palace on the Adriatic coast to tend his vegetable garden. Later he was asked to come back and help sort out the empire. His reply, as an ancient summary of the emperors’ lives records it: “If you could see the cabbages I have planted here with my own hands, you surely would never have thought to request this.” It remains the greatest out-of-office reply ever written.

Why is the slow version the dangerous version?

If any one emperor had cut the silver from 98 percent to 5 in a single announcement, there would have been a riot, and the coin would have been refused at every market stall in the empire.

That isn’t what happened. Picture a family instead. A grandfather saves his pay in good denarii. His son inherits a purse of coins that ring a little duller. His grandson inherits coins that don’t ring at all, and since he can’t remember them ringing, he doesn’t miss the sound. Nobody in that family lived through a crisis. Each of them lived through a slightly worse coin, which feels very different and costs about the same.

Wherever two coins share a face value, people spend the worse one and keep the better, a habit reliable enough to have its own name, Gresham’s law. What nobody could do was get out of the denarius itself, since wages and prices were counted in it. Each step down had a good excuse: a war, a plague, a rival claimant, an army expecting to be paid by Friday. No one of those decisions destroyed the denarius. All of them did. Debasement needs no villain. It needs an institution with a habit, and habits outlive emperors.

What does Rome teach us about money?

Rome could do almost anything to a coin. It could shrink the coin. It could not shrink the ounce.

A denarius was whatever the mint said it was this year. A gram of silver was a gram under Augustus, a gram under Gallienus, and it’s a gram this afternoon. Anyone who held metal by weight watched the coins melt away around them. Anyone who held coins by decree discovered that the decree was all they had ever owned. The buying power didn’t evaporate. It moved, from the people holding the tokens to the authority issuing them, and that move is what debasement is.

Now the bear case for our own position, in plain words. Silver is volatile, pays nothing and costs money to store. Historians still argue over how much of Rome’s slide was policy and how much was desperation.

All true, and notice which complaint is missing. A swinging price is a quarrel with the market. A diluted coin is a quarrel with the issuer, who also writes the rules, and in 301 the rules included the death penalty. Silver has no issuer. No emperor could print it, water it down or declare it a double, and it outlasted every one of them. Rome is an argument about the unit, not a promise about the price.

We have one advantage the Romans lacked. They couldn’t see their coin’s silver plotted on a chart; they felt it over generations, in the price of bread. We measure ours monthly, to the decimal, and argue about the methodology. That advantage is worth exactly nothing unless somebody looks.

The pot from Frome sits in the Museum of Somerset in Taunton now: 52,503 coins, every one still wearing an emperor and a promise, the newest of them struck around the year the lion-pricer went home to his garden. Archaeologists suspect whoever buried it never meant to come back, and that it may have been an offering for a good harvest. If so, the gods received four years of a soldier’s pay. Like every tip jar since, it looked generous until somebody counted it: mostly copper, a little silver paint, and the emperors had already taken their cut.

Sources

• Silver content by emperor (Augustus, Nero, Trajan, Severus, Gallienus) and the antoninianus tariffed at twice a denarius while weighing about 1.5 times as much, with less than 70% of the silver of two denarii: NGC Ancients, Roman silver coinage.
• Weights and purity of the denarius under Augustus and Nero, the antoninianus at 5.1 grams and 52% fine, and the roughly 5% purity by AD 300: Denarius, Wikipedia (a secondary source).
• The Edict on Maximum Prices of AD 301, its death penalty for profiteers, its uneven enforcement and its decline by 305: Edict on Maximum Prices, Wikipedia (a secondary source).
• The figures above are approximate and vary somewhat between sources; the ~30% shortfall in the antoninianus is our own arithmetic from the NGC figure. So are the $6.64 (3.9 grams at 98% silver, at $54.05 an ounce, silver’s price on October 24, 2026 from gold-api.com) and the milk jug (a 128-ounce gallon at 5% of the original 98%, about 6.5 ounces).

Questions people actually ask

What was Roman currency debasement?

Roman debasement was the gradual reduction of the silver in Roman coins while their name and face value stayed the same. The denarius fell from roughly 98% silver under Augustus to about 5% by AD 300. Governments gained spending power while holders of the coins lost purchasing power.

How much silver was in a Roman denarius?

It varied by period. Under Augustus a denarius was about 97.5 to 98% silver and 3.9 grams. By Nero in AD 64 it was about 93.5%, by Trajan roughly 89 to 90%, and by about AD 300 around 5%.

Did Rome invent shrinkflation?

Rome did not invent shrinkflation, but its coinage is an ancient version of the same trick: keep the name, price and appearance while reducing the substance. The difference is that Rome applied it to the money itself, not to a product sold in exchange for money.

What was the antoninianus?

Introduced by Caracalla in AD 215, the antoninianus was a coin worth two denarii that weighed only about 1.5 times as much. It held less than 70% of the silver in two denarii, which made it a quiet cut in value dressed up as a bigger coin.

Why did Rome debase its currency?

Emperors faced costs they could not easily raise taxes to cover, including wars, plagues, rival claimants and armies that expected pay. Reducing the silver in each coin let the mint produce more coins from the same metal. Historians still debate how much was policy and how much was desperation.

What was Diocletian's Edict on Maximum Prices?

Issued in AD 301, the edict set legal price ceilings on well over a thousand products and services and prescribed the death penalty for profiteers. It was patchily enforced and was largely no longer in force by the time Diocletian abdicated in 305.

Did Roman price controls work?

Not lasting. The edict showed some short-term effect, but its price limits were not consistently observed and it was weakly enforced in the western empire. Scholars generally describe it as of limited success against inflation, since it treated rising prices instead of the debased coinage.

What does Roman debasement teach us about modern money?

That a unit defined by decree can lose purchasing power slowly enough that no single step causes alarm. A gram of silver stayed a gram under every emperor, while the denarius changed with each reform. Rome is a lesson about the unit, not a prediction about prices.

When did the Roman denarius stop being silver?

It never officially stopped; it was diluted. The denarius fell from roughly 98% silver under Augustus to about 57% under Septimius Severus, and by around AD 300 it was about 5% silver. Under Gallienus the antoninianus was a copper coin with a thin silver wash, at 5% or less.

Why do some Roman silver coins look copper colored?

Because by the reign of Gallienus many were copper coins with only a thin wash of silver on the surface, at 5% silver or less. The wash wore off with handling, so a coin that started shiny turned brown as the copper showed through. The coin, in effect, revealed its own debasement.

What did Lactantius say about Diocletian's price controls?

Lactantius, a Christian writer hostile to Diocletian, wrote that after the 301 edict much blood was shed over trifles, merchants were afraid to offer goods for sale, and scarcity grew worse until the law was abandoned. Historians treat his account with caution but agree the edict had limited success against inflation.

What was the Frome Hoard?

A clay pot of 52,503 Roman coins found by a metal detectorist near Frome, England, on April 11, 2010. The coins weighed about 350 pounds, date from AD 253 to 305, and are almost all heavily debased bronze rather than silver. In its day the hoard was worth about four years of a Roman soldier's pay. It is now in the Museum of Somerset in Taunton.