Introduction[2]
The name econometrics was a neologism created by Ragnar Frisch to characterise a discipline concerned with advancing economic theory in its relation to statistics and mathematics.
As a founding member of the Econometric Society and its journal Econometrica in the early 1930s, Frisch wanted to promote research that unified ‘the theoretical-quantitative and the empirical- quantitative approach to economic problems’ (Frisch 1933: 1). Since then, however, the term econometrics has come to signify just the statistical aspects of quantitative economics research as with A Textbook of Econometrics (Klein 1953) or just Econometrics (as in Valavanis 1959). There remained a branch emphasising the more general aspect, in that textbooks were titled Statistical Methods of Econometrics (see Malinvaud 1966), which was also the name of the main econometrics course for the Master of Philosophy (MPhil) degree in Economics at the University of Oxford when the first author arrived there in 1982.1 We are grateful to Steve Bond, John Creedy, Christopher L. Gilbert, Grayham E. Mizon, James Poterba and Jan Toporowski for helpful information about Oxford econometrics and recollections from their time at the University and to John Gittins for permission to quote from his history of Oxford statistics (Gittins 2013).
Given the relatively recent definition of econometrics, and its subsequent narrowing, much of the early history of the discipline must be conceptualised as economics in relation to statistics, as we do here, including the creation and curation of observations on economic phenomena. Moreover, “statistics” still refers both to the discipline which studies methods of statistical analysis (as in a department of statistics) and to summary measures of observations (as in the statistics of crime). Schumpeter (1933: 5) claimed that in contrast to the physical sciences that had to create their measures, ‘Some of the most fundamental economic facts, on the contrary, already present themselves to our observation as quantities made numerical by life itself'. However, that still requires that such facts be recorded and combined over events, time and people to be useful for analysis.
We include researchers who undertake such invaluable tasks as econometricians, which leads to a surprisingly rich history of the subject at Oxford before the 1930s.General histories of econometrics are provided by Morgan (1990) and Qin (1993, 2013), with an overview and selected reprints of key papers in The Foundations of Econometric Analysis by Hendry and Morgan (1995). Thomas (2018) records the important role the London School of Economics (LSE) played in the development of econometrics in the twentieth century and also in beginning the study of the history of econometrics. Oxford econometricians continued that development of the history of their discipline. In addition to the two books by Qin and that by Hendry and Morgan, see Qin and Gilbert (2001) and Gilbert and Qin (2006, 2007), both of whom had been doctoral students or faculty at Oxford.
The structure of this chapter is as follows. Section 2 describes the early history of contributions to quantitative economics as embryonic econometrics from the seventeenth to the nineteenth centuries. Section 3 discusses econometrics at Oxford over 1900-1980, including major advances in creating aggregate economic measurements. Section 4 updates the history from 1980 to 2000, before Section 5 records Oxford econometrics in the twenty-first century to about 2010, after which point it is no longer “history”, although such time divisions are arbitrary and many individuals span several of these sections. Section 6 describes contributions to data provision in the twenty- first century before Section 7 considers the most recent addition of climate econometrics, developing and applying econometric tools for analysing climate data, which is driven by human economic behaviour and so faces much the same slew of econometric problems as macroeconomic time series.
2