Econometrics: A Biography of the Concept
A monograph on the concept of econometrics — from its Greek roots (oíkos, nómos, métron), through Paweł Ciompa's Lwów 'econometry' (1910) and Ragnar Frisch's definition (1926), the institutionalisation (Econometric Society 1930, Econometrica 1933), Haavelmo's probabilistic revolution (1944) and Oskar Lange's Polish school (1958), to contemporary causal machine learning.
From the Lwów “econometry” to the era of causal machine learning.
Before turning to methods, it is worth knowing the concept itself. The history of the word econometrics — from the Greek oíkos to the algorithms of artificial intelligence — mirrors the path of the whole discipline: from the management of a household to the rigorous, quantitative measurement of economic processes. This article is a monograph on that concept: its etymology, birth and evolution.
Chapter I. The philological foundation of the neologism
The word econometrics (French économétrie, Polish ekonometria) is not merely the name of a discipline but a precisely constructed, hybrid-compound neologism whose roots lie in classical Greek. To grasp its weight, it must be broken into three elementary components.
- oiko- — from the Greek oíkos ($\text{οἶκος}$): house, household, estate, dwelling.
- -nomia — from nómos ($\text{νόμος}$): law, rule; and the verb némein ($\text{νέμειν}$): “to manage”, “to distribute”.
- -metria — from métron ($\text{μέτρον}$): measure, measurement, due proportion.
The compound oikonomía ($\text{οἰκονομία}$) in ancient Greece — for instance in Aristotle (Politics, Book I) — originally denoted “the management of the household”. In modern scientific terminology the suffix -metria — by analogy with geometry (the measurement of the earth) — designates disciplines that join the theoretical basis of a science to its quantitative, empirical measurement. Econometrics is therefore, literally, “the measurement of the economy”.
Chapter II. The Lwów precursor — Paweł Ciompa (1910)
Although the worldwide fame of econometrics is tied to the year 1926, the term first appeared in the literature in 1910, through the Polish banker and accounting scholar Paweł Ciompa (1867–1913).
Ciompa, then secretary of the board of the Commercial School Society in Lwów, published two key works:
- in Polish: Zarys ekonometryi i teorya buchalteryi (“An Outline of Econometrics and the Theory of Bookkeeping”; Ciompa 1910a);
- in German: Grundrisse einer Oeconometrie… (Ciompa 1910b).
Ciompa’s econometrics was descriptive and geometric in character. He proposed that economic phenomena be represented — on the model of physics — through “econographics”. He defined the value of a good as the product of quantity and price, which corresponded geometrically to the area of a rectangle. On this view, bookkeeping (accounting) was the practical application of econometrics. Although this conception differed from today’s mathematical-statistical econometrics, Ciompa is recognised in the international literature as the precursor of the name (Gruszczyński 2022).
Chapter III. The Norwegian revolution — Ragnar Frisch (1926)
The modern, mathematical-statistical meaning of econometrics was given by the Norwegian economist Ragnar Frisch (1895–1973), later the first laureate of the Nobel Prize in Economics (1969). He first used the term économétrie in his French-language essay Sur un problème d’économie pure (Frisch 1926).
Frisch, inspired by theoretical physics, argued that economics must become an exact science. He defined econometrics as a synthesis of three elements: economic theory, mathematics and statistics. He stressed that statistics without theory is blind, while theory without numerical verification is barren. Notably, Frisch later admitted that he had been unaware of Ciompa’s works when he coined his term (Bjerkholt 2017) — a case of the same name being independently minted twice.
Chapter IV. Institutionalisation — the Econometric Society (1930)
The concept of econometrics was finally sanctioned on 29 December 1930 in Cleveland, when — on the initiative of Irving Fisher, Ragnar Frisch and Charles Roos — The Econometric Society was founded (Bjerkholt 2017).
The key moments of this stage:
- 1933 — the launch of the journal Econometrica (Frisch 1933), which remains to this day the foremost publication in the field.
- The English form — the addition of the suffix -ics to econometric (hence econometrics), emphasising the systematic character of the discipline, by analogy with mathematics or physics.
Chapter V. The probabilistic revolution — Trygve Haavelmo (1944)
Early econometrics was largely deterministic. The true breakthrough came in 1944, when Frisch’s student Trygve Haavelmo published The Probability Approach in Econometrics (Haavelmo 1944).
Haavelmo brought the calculus of probability into econometrics, showing that simplified economic theories can be rigorously tested only when disturbances are formalised as random variables. It is this current that ties econometrics to the method of least squares and the central limit theorem. For his contribution Haavelmo received the Nobel Prize in 1989.
Chapter VI. Evolution in Polish — Oskar Lange (1958)
In Polish, the term underwent an orthographic evolution: from Ciompa’s “ekonometrya” (consistent with the older spelling of “geometrya”) to the modern “ekonometria”. By using the ending -trya, Ciompa built a direct bridge between the geometric structure and the newly conceived discipline of measuring value.
After the Second World War the foundations of the modern Polish school were laid by Oskar Lange. His textbook Wstęp do ekonometrii (“Introduction to Econometrics”; Lange 1958) integrated the statistical apparatus with cybernetics and the theory of optimal programming, becoming a work translated into many languages and educating generations of Polish economists.
Chapter VII. The present — causal machine learning (2018)
Today, more than 110 years after Ciompa, econometrics is undergoing another renaissance through artificial intelligence. In 2018 Victor Chernozhukov and co-authors introduced the method of Double/Debiased Machine Learning (DML) (Chernozhukov et al. 2018).
Contemporary causal econometrics (causal machine learning) uses random forests and neural networks to measure causal effects precisely — for instance, the impact of a new technology on productivity — in large data sets (big data), while preserving the rigour of inference that Frisch demanded.
A chronology of the term
| Year | Event |
|---|---|
| 1910 | Paweł Ciompa — the birth of the word “ekonometrya” in Lwów |
| 1926 | Ragnar Frisch — the scientific definition of “économétrie” |
| 1930 | Fisher, Frisch, Roos — the founding of The Econometric Society |
| 1933 | First issue of the journal Econometrica |
| 1944 | Trygve Haavelmo — the probabilistic revolution (Nobel 1989) |
| 1958 | Oskar Lange — the consolidation of the Polish school |
| 2018 | Victor Chernozhukov et al. — the era of Double Machine Learning |
Econometrics, despite its changing tools — from Ciompa’s rectangles to AI algorithms — has invariably pursued the same goal: turning economics into an exact science through measurement.
References
Aristotle. Politics, Book I. Translated by C. D. C. Reeve. Indianapolis: Hackett.
Bjerkholt, Olav. 2017. “On the Founding of the Econometric Society.” Journal of the History of Economic Thought 39 (2): 175–198.
Chernozhukov, Victor, Denis Chetverikov, Mert Demirer, Esther Duflo, Christian Hansen, Whitney Newey, and James Robins. 2018. “Double/Debiased Machine Learning for Treatment and Structural Parameters.” The Econometrics Journal 21 (1): C1–C68. https://doi.org/10.1111/ectj.12097.
Ciompa, Paweł. 1910a. Zarys ekonometryi i teorya buchalteryi. Lwów: Towarzystwo Szkoły Handlowej.
Ciompa, Paweł. 1910b. Grundrisse einer Oeconometrie und die auf der Nationalökonomie aufgebaute natürliche Theorie der Buchhaltung. Lemberg–Kraków.
Frisch, Ragnar. 1926. “Sur un problème d’économie pure.” Norsk Matematisk Forenings Skrifter, ser. 1, no. 16: 1–40.
Frisch, Ragnar. 1933. “Editorial.” Econometrica 1 (1): 1–4.
Gruszczyński, Marek. 2022. “Accounting and Econometrics: From Paul Ciompa to Contemporary Research.” Journal of Risk and Financial Management 15 (11): 510. https://doi.org/10.3390/jrfm15110510.
Haavelmo, Trygve. 1944. “The Probability Approach in Econometrics.” Econometrica 12 (Supplement): iii–vi, 1–115. https://doi.org/10.2307/1906935.
Lange, Oskar. 1958. Wstęp do ekonometrii [Introduction to Econometrics]. Warsaw: Państwowe Wydawnictwo Naukowe.
Next: The method of least squares — a complete treatise · The normal distribution · The central limit theorem