Exploring the History of Statistical Inference in Economics

Exploring the History of Statistical Inference in Economics. 2021.  Edited by Jeff Biddle and Marcel Boumans. Supplement to volume 53 of HOPE. Durham, NC: Duke University Press.

"Exploring the History of Statistical Inference in Economics: Introduction," by Jeff Biddle and Marcel Boumans (pp. 1–24). Attention to inference as an analytically separable aspect of research involving statistical data can be a fruitful perspective from which to understand the history of empirical economics.

Inferences in the Field

"Statistical Parables: Agricultural Economists, Development Discourse, and Paths to Modernization during the Cold War," by Paul Burnett (pp. 27–52). Theodore Schultz's use of statistical case studies or "statistical parables" were normative and even moral exemplars for shaping economic and social policy

"Statistical Inference in Economics in the 1920s and 1930s: The Crop and Livestock Forecasts of the US Department of Agriculture," by Jeff Biddle (pp. 53–80). The problem of creating good crop and livestock forecasts was, for the Bureau of Agricultural Economics, almost entirely a problem of applied statistics, involving little economic theory.

"False Accounting as Formalizing Practices: The Computation of Macroeconomic Aggregates in African Countries since Structural Adjustment," by Boris Samuel (pp. 81–110). IMF practices in Mauritania in the 2000s led to what can be called "false accounting," favoring conceptions of consistency that neglect basic requirements of macroeconomic analysis.

Inference in Time

"Narrative Influence with and without Statistics: Making Sense of Economic Cycles with Malthus and Kondratiev," by Mary S. Morgan (pp. 113–138). Inference in the nineteenth century involved a variety of flexible practices according to the questions and materials at issue, with one practice being "narrative making" to explore and explain statistical information.

"Searching for a Tide Table for Business: Interwar Conceptions of Statistical Inference in Business Forecasting," by Laetitia Lenel (pp. 139–174). Statistical inference with probability was not the long-sought solution for the problem of objectivity but a long-contested, and repeatedly discarded, approach.

"Revisiting the Past? Big Data, Interwar Statistical Economics, and the Long History of Statistical Inference in the United States," by Thomas A. Stapleford (pp. 175–203). The parallels between the approaches to economics that have accompanied Big Data and the vision for statistical economics outlined by Wesley Mitchell in 1924 complicate any simple, linear narrative about the history of statistical inference in economics

Inference without a Cause

"Pictorial Statistics," by Marcel Boumans (pp. 207–226). Although Francis Galton aimed for mechanical objectivity, subjective judgments nevertheless appear to be necessary for this kind of inductive inference to achieve purity.

"Making Inferences from Index Numbers (1860–1914)," by Aashish Velkar (pp. 227–258). By examining the early history of index numbers, it becomes clear that the particular context--who made the inference and why--is especially pertinent, making inference a cognitive and not just a heuristic process of reasoning.

"The Case against 'Indirect' Statistical Inference: Wassily Leontief's 'Direct Induction' and The Structure of American Economy, 1919–29," by Amanar Akhabbar (pp. 259–292). Leontief conceived his interindustry method as an alternative to the inferential strategies of interwar econometrics and statistical economics.

"Politicizing the Environment: (In)direct Inference, Rationality, and the Credibility of the Contingent Valuation Method," by Harro Maas (pp. 292–323). Contingent evaluation went from being a rather innocuous method used in cost-benefit analysis to a politically and judicially charged method for value claims.