Roman Real Wages
When first coming to grips with the “grassroots” or top-down consumers of the Roman economy, there’s hardly any better way to measure their individual economic prosperity than by taking their real wages into account.
The Roman socioeconomic historian Walter Scheidel defines real wages as “the value of wage income relative to that of consumer goods” (Scheidel 2010). In other words, a real income is the amount of goods and/or services one can purchase for what one’s cash is worth at any given point in time and space.
The historian Robert Allen is renowned for his work on real incomes in early modern Europe. He has also published research pertaining to Roman real wages, drawing from the invaluable Diocletianic price edict of 301 CE.
Allen’s approach to this field of study includes a particular conceptual framework he uses revolving around a “consumption basket”, which is the entire bundle of goods and services that a person regularly consumes. The two kinds of consumption baskets are ‘baskets of respectability’ and ‘bare bones baskets’.
Baskets of respectability were fit for skilled workers’ consumption, meanwhile the latter, although having the same calorie counts, provided them with a less diverse and desirable diet. That’s of course coupled with minimal non-food resources for the poorest citizens of the empire already living on bare subsistence at the time. According to Allen, a Roman unskilled worker or operarius could afford 56% of a respectable basket and 110% of a bare bones basket.
Scheidel’s own estimates point more in the other direction, downsizing them to 30% and 80% for an unskilled rural labourer, respectively. It’s worth noting that the real income of Roman Egyptians in the second and third centuries CE, defined in terms of bare bones consumption baskets for unskilled workers, was a milestone away from reaching the levels of, say, Amsterdam and Delhi, 1 compared to 4.7 and 3 (see the first graph in the article listed below).
That’s not to create a black-and-white picture of the ancient world: Neobabylonian-Achaemenid Mesopotamia experienced an elevation in real wages in the sixth century BCE, possibly as a result of demographic dislocations and contractions in the seventh century.
In terms of economic performance, however, it’s more convenient to use wheat wages as a measurement for comparing national real incomes cross-culturally and geographically over larger swathes of time. Had it not been for this, the historical study would’ve been restricted to mere comparative analysis between economically well-attested societies.
Grain prices are also easier to come by than aggregates such as Diocletian’s aforementioned Edict on Maximum Prices. One does this by converting income and a variety of goods into units of wheat.
Classical Athens had the highest daily wages in wheat equivalents for unskilled labourers in the fifth and fourth centuries BCE, unparalleled until modernity. An annual salary of a labourer standing at around 1,043 liters was greater than the real GDP per capita of a Greek person between 1833 and 1860, fluctuating between 520 and 730 liters of wheat.
The case of economic growth in Greece in premodern economics is unique, warranting more respect than to be cast aside as a poor agrarian economy (for the commoners) compared to the Persian upper class, as Herodotus wants us to believe.
Athenian real wages were also higher than the Roman level, referring to conditions in Hellenistic Delos in the third century BCE. The annual wheat wage was 2,200–3,900 liters in Athens, compared to 1,200 liters for an unskilled rural labourer (Scheidel) and 1,400 liters in an annual free wage (Jongman) in the Roman Empire.
Although low real wages could have been widespread around the Mediterranean, it is quite impossible to say so without laying serious credence to extrapolation and conjecture, as historical records are scant and maldistributed in terms of time and place; generally, the Diocletianic price edict and Egyptian papyri are the best resources we have at our disposal.
Scheidel touches upon two commonly attributed factors of equalizing real income: economic growth and population loss.
There’s currently no evidence to support the assertion that economic growth elevated real wages during the Roman period as opposed to population regression. In spite of this, Scheidel doesn’t discredit it entirely due to “absence of evidence” not being “evidence of absence”.
What we do know, is that the Black Death wreaked havoc on the demographic situation leading to the deaths of at least one third of Europeans, maybe even half of them. As labour became less available, workers saw their bid to demand higher wages and better conditions from their feudal lords, which in turn resulted in higher real incomes.
If they didn’t comply, they would simply pack their bags and move to where work was in demand, which was anywhere, often the cities; the feudal system had got into an irrevocable mess. Most daily wheat wages 3,000 years before the bubonic plague occurred varied from 5 (+/-30%) liters of wheat equivalents.
Widening income inequality and an unfavourable ratio of land to labour were the norm for premodern economies, and the major equalizers identified for the first one by Scheidel are “transformative revolutions”, mass-mobilization warfare, plague and state collapse. To create meaningful economic inequality, a juggernaut of uncomfortable societal upheaval must be instigated; usually, government policy doesn’t cut it.
Following the Antonine plague of the 160s, labour costs (wages; draught animals) rose relative to the price of wheat, more so for non-essential goods. Land values (prices; rents) fell precipitously. By reducing the labour supply, plague leaves the surviving workers, unskilled and skilled, with increased real incomes. A similar development is supposed to have happened during the Justinianic plague, a series of pandemics striking Western Eurasia from 540–ca 750 CE.
A few more enticing possibilities for engaging with the topic is to consider the connection between real wages and slave labour and physical well-being (body height).
The first one is on no steadfast ground due to a lack of representative slave prices and reliable civilian wage statistics from late Republican Italy, where academic sources on slavery are the best overall, tied with Roman Egypt.
The second fact of height supports a Malthusian angle on the Roman economy, as greater population density and population-> more skewed ratio of land to labour -> depressed real wages -> poorer nutrition and deleterious epidemiological conditions -> lower stature.
A falling population, conversely, contributes to greater real incomes and heights: both stature and real wages rose concurrently after equalizing societal collapses and plagues (the fall of antiquity, the Black Death) came into fruition. Heights began to rise first when the Roman Empire unraveled, primarily in the fourth century CE.
Naturally, there’s no definite correlation between real wages and stature, so inferred estimates must, as usual, be taken with a grain of salt. The same applies for extrapolation and conjecture from biological and demographic facts as well.
In the case of Greece, real incomes are thought to have risen with population growth, the former being attributed to uncommon institutional arrangements.
The historian and political scientist Josiah Ober explains in The Rise and Fall of Classical Greece that arrangements such as collective-self governance, civic rights, the rule of law and protection from theft of your hard work etc. spurred economic growth, rational trust, cooperative knowledge-sharing, investment and risk-taking that in turn improved wages and thus real wages.
Further reading:
“Roman real wages in context” Walter Scheidel, Stanford University (2010)
“Human development and quality of life in the long run: the case of Greece” Walter Scheidel, Stanford University (2010)
Ober, Josiah (2016). The Rise and Fall of Classical Greece, Princeton: Princeton University Press.
