Appearance
Cliometrics is the use of economic theory and rigorous statistics to answer historical questions — measuring wages, prices, output, and living standards, and testing explicit models against the evidence rather than relying on narrative alone. To get started you need a well-defined question with a measurable outcome, a defensible long-run dataset, a tool such as R or Stata, and the discipline to state your model and its assumptions openly. This guide walks the workflow end to end.
What is cliometrics, in one paragraph?
The word fuses Clio, the muse of history, with "metrics". It emerged in the 1960s when economists like Robert Fogel and Douglass North began applying formal models and regression to the past, work that won them the 1993 Nobel Memorial Prize. The defining moves are quantification, an explicit model of behaviour, and statistical testing — including, famously, counterfactual reasoning about what would have happened under different conditions.
How does it differ from ordinary economic history?
| Narrative economic history | Cliometrics | |
|---|---|---|
| Evidence | Documents, qualitative | Measured variables |
| Argument | Interpretive | Model plus statistics |
| Outcome | A persuasive account | A tested, falsifiable estimate |
| Typical claim | "Trade grew rapidly" | "Trade grew 1.8% per year, p < 0.01" |
Both are legitimate; cliometrics simply commits to numbers and explicit assumptions so its claims can be checked and contested.
What is the practical workflow?
- Frame a measurable question. "Did real wages rise during industrialisation?" forces you to define real wages and a period.
- Assemble a long-run series. Combine wage books, price indices, and population estimates into a tidy table.
- Deflate and index. Convert nominal figures to real terms with a price index so a 1780 shilling and an 1850 shilling are comparable.
- Model and test. Fit a trend or regression, and consider the counterfactual.
- Report with uncertainty, stating assumptions and the fragility of the sources.
r
library(dplyr)
wages <- read.csv("artisan_wages.csv") |>
mutate(
real_wage = nominal_wage / price_index * 100, # base year = 100
log_rw = log(real_wage)
)
# average annual growth in real wages
fit <- lm(log_rw ~ year, data = wages)
coef(fit)["year"] * 100 # ~ % growth per yearWhy do counterfactuals matter?
Cliometrics is unusual in modelling the road not taken. Fogel's railways study estimated the "social saving" — how much cheaper railways made transport than the next-best alternative — by constructing a counterfactual economy reliant on canals and roads. His provocative conclusion was that the saving was modest, challenging the idea that railways were indispensable. Whether or not you accept it, the method forces explicit reasoning about causation.
Where do I find the data?
Much of the spadework is already shared:
- Maddison Project — long-run GDP per capita across countries.
- Clio Infra — global datasets on inequality, health, and human capital.
- Allen's wage and price series — building-worker wages and a welfare-ratio framework.
- National statistical reconstructions — for example, Broadberry et al. on British GDP back to 1270.
Always read the source notes; these series are themselves reconstructions with assumptions you inherit.
What criticisms should I take seriously?
Sceptics, including some of cliometrics' own founders later in life, warn that models can smuggle modern rationality onto past actors, that counterfactuals are inherently speculative, and that the measurable can crowd out the meaningful. The remedy is not to abandon numbers but to pair them with source criticism, report uncertainty honestly, and treat estimates as contributions to debate rather than verdicts.
Key Takeaways
- Cliometrics applies economic theory and statistics to historical questions with explicit models.
- It differs from narrative economic history by committing to measured, testable claims.
- The workflow: frame a measurable question, build a long-run series, deflate to real terms, model, report uncertainty.
- Counterfactual reasoning, as in Fogel's railways study, is a signature method.
- Shared datasets like the Maddison Project and Clio Infra give you a running start.
- Guard against imposing modern assumptions, and always pair numbers with source criticism.
Frequently Asked Questions
What is cliometrics?
Cliometrics is the application of economic theory and formal statistical methods to historical questions, especially economic and social history. It pairs quantitative analysis with explicit models to explain why past economies and societies behaved as they did.
How is cliometrics different from economic history?
Economic history is the broad field; cliometrics is the quantitative, theory-driven strand within it. Cliometrics insists on measurable variables, explicit models, and statistical testing rather than narrative alone.
What was the famous counterfactual in cliometrics?
Robert Fogel's study of railways estimated the 'social saving' by modelling a counterfactual America without them, arguing the economy would have grown almost as much via canals and roads. It helped define the field and won a Nobel prize in 1993.
What skills do I need to start?
You need basic statistics, comfort with a tool like R or Stata, and the historian's craft of source criticism. The hardest skill is judging whether your sources can bear the quantitative weight you put on them.
What data do cliometricians use?
Wages, prices, output, population, trade, and tax records are staples, often drawn from censuses, account books, and price series. Much sits in shared databases like the Maddison Project and Clio Infra.
What are the main criticisms of cliometrics?
Critics warn that models can impose modern assumptions on past actors, that counterfactuals are speculative, and that quantifiable sources crowd out unmeasurable but important factors. Good practice meets these head-on.