So how much were Roman coins worth in their day?
It's a surprisingly difficult question to answer with any degree of certainty but it comes up often enough. The main problem is in trying to equate the buying power of ancient Rome with modern money. A dollar that buys a song online can't be compared with coins used to buy slaves and rotting eels. Before we look at the available data there is another complicating factor in that inflation and the passage of time distorts the understanding of the purchasing power of the various denominations. Just, again, as a Civil War-era dollar doesn't have the same purchasing power as a dollar does now.
We can nevertheless come up with some approximations. Some things, and sevices, are more or less timeless. Buying a loaf of bread or a haircut now or two thousand years ago is more or less the same experience and can at least be roughly compared to in terms of cost and effort to "produce". If we knew the average Roman's paycheck and what some of these basic staples cost back then, all for a given point in time, we could extrapolate by referencing how many haircuts or bread loaves the average person can buy today given a typical salary now. We don't, unfortunately, have the full picture but we can come up with enough data to make some reasonable guesses.
The first century provides us with the most in-depth historical snapshots of Roman life thanks to Suetonius, Pliny and Cassius Dio. Suetonius writing around the year 100, for example, gives us some tantalizing tidbits. He notes that Domitian just a few years before had raised the annual salary of a common soldier by 25% from 9 to 12 Aurei. Since it's obvious that soldiers weren't paid nine (or twelve) coins in a year right there that tells us that he was talking the same way we would mention an employee drawing fifty or a hundred thousand dollars: the yearly tally rather than the actual paycheck. Knowing that the Sestertius was valued at 1/100th the value of an Aureus it becomes much more likely that a soldier's pay was given out in this denomination which would have been far more useful in the everyday commerce a soldier would have engaged in and, at let's use the handy new payscale, 1200 Sesterces works out to an even 100 monthly.
Those soldiers however had to pay "taxes" in the form of deductions meant to offset the cost of room and board, training and their gear, none of which was free, so that the actual take-home amount was significantly less. How much their real pay, above and beyond the costs they were required to reimburse the state, is somewhat vague; perhaps half their gross had to be turned over for these off-the-top expenses. On the other hand though we can assume that they supplemented their income in any imaginable number of ways loosely controlled weaponized authority figures could get away with. So, for a working figure, let's assume that after all was said and done that this soldier managed a clean 100 Sesterces a month and this figure being good for a generation either side of the year 100 AD.
The non-commissioned soldier of Domitian's time was regarded as a blue collar earner. A state pension for successful retirement following a 25-year career, the statistical chances of whose completion must have been dismally small, could have been a distant dream new enrollees looked forward to since it was valued at a pound of gold. That works out to a little over $20,000 today... a figure that while seemingly very humble for such dedicated service must nevertheless be taken into the context that this ancient Roman likely had very humble expenses to begin with. Twenty g's buys a lot of cheap wine and wheat grain after all and you may consider that most people in the third world today, whose buying needs much more closely parallel those of an ordinary Roman, get by without ever seeing a similar lump sum payout.
So assuming this yardstick of 100 Sesterces a month were to be considered a decent salary for a typical worker of low- to medium-skill this works out to somewhere between three and four of these big coins with which to carry the day for a proverbial family of four - or we can round up to an even four, a Denarius, just for the sake of round numbers. With the average salary for someone in this skill-range in the U.S. hovering around $40,000 a year (let's assume a net of $32,000 after taxes or around $90 a day) it's tempting to draw a quick analogy to conclude that a Denarius had back then the same buying power as $90 does today. But is this even roughly accurate?
In order to answer that I've taken some assumptions that may not hold true and, in any case, we have yet to list the costs of any goods during Roman times. Let's do a little stress testing and see how our figures are holding up. Fortunately, one missing piece of the puzzle is not only known but is also contemporary to this era: the Pompeii disaster locked in pricing tables for a number of generic commodities current to the Autumn of 79. In it we learn a particularly useful price for bread which, again, we can take as a good reference point of value across the ages since a handmade loaf must have been just about as hard and costly to make back then as it is now. That price was one As, a sixteenth of a Denarius. Does then a loaf of artisan (handmade) bread today run about $90/16=$5.60? I'd say this is spookily close to shopping in the Whole Foods bakery section. Really not a bad stab initial conclusion. It is even more impressive if we take that price and transpose it to a country like Zambia or Bolivia where it would run maybe a dollar and where the labor of a baker is likely much more commensurate with that of an ancient Roman. But, either way, this workup is perhaps a lucky strike and we really need some more numbers before we feel confident about these valuations.
Just after Domitian comes Nerva who Suetonius tells us started a program to help orphaned children. A state stipend of 16 monthly Sesterces for boys and 12 for girls would be allotted by the state. Assuming that one of these kids would have depended entirely on this stipend that works out to a daily Dupondius for him and about an As-and-a-half for her which leaves them to fend for themselves with a pound of bread and a few pennies for other expenses on which to live off. Considering that they would in all likelihood supplement this with handouts, odd jobs and maybe a little thieving here and there, the way they would have gotten by just prior to Nerva's munificence, for an indigent's standards this does not sound bad at all. Taking our 5-and-change computation on the worth of an As this would work out to about a thousand dollars a year free from uncle Nerva. And that's surely more than kids in a Mexican slum make per annum.
In the end are we comfortable saying that a Denarius should be worth about $90? It has been estimated at less. Someone once came up with $20 using similar methods and while not very generous the figure still works (that artisan bread now costing a measly $1.25 which is still accurate if we shop for it at Costco (or a third world country!)). Pompeii also gives us that a tall glass of wine, the vintage 79 sort suitable for a not-too-discriminating lunch, also costs an As. Again, whether we use the lower or the higher figure this is still within the range of acceptable even if in either case we envision booze fit only for bums.
In the end, we can say that an As was very likely, at least during the time of the Flavians, the denomination that would have been most useful at the market while a Denarius - a good day's pay for a chunk of the workforce - certainly packed enough punch to put plenty of food on the table for a family's meal. It also means that a careful budget placed the average Roman in a bit of a bind since the choice of eating well and having little left over for other expenses would mean uncomfortable sacrifices. And when it comes down to it, whether you're supporting a family in the states on $90 a day (or $20 a day in South America) you're not going to exactly live large either.