SOPhiA 2018

Salzburgiense Concilium Omnibus Philosophis Analyticis

SOPhiA ToolsDE-pageEN-page

Programme - Talk

The Presuppositions Behind Causal Inferences in Economics
(Philosophy of Sciecne, English)

Claveau and Mireles-Flores (2014) analyzed a case study (an OECD report on causes of unemployment) with the aim of describing presuppositions on causality held by economists. Their article concludes indicating that causal inferences in economics are relatively poorly researched and further research is needed. For instance, the question which of the philosophical theories of causality is held as true (Claveaou and Mireles-Flores 2014). Maziarz (2017a) reviewed the philosophy-of-economics research on causality and indicated that, with few exceptions, the methodologists focus on conducting normative analyses. Contrary to the hopes of economic methodologists, such normative considerations have an insufficient influence on the practice of economists (Hutchison 2000). In fact, the debate whether the mainstream-economics models should be interpreted causally or, in contrary, operate by means of functional dependencies is far from being settled (cf. Verreault-Julien 2017). The recently widely discussed "Growth in a Time of Debt" (Reinhart and Rogoff 2010) controversy showed that grounding economic policy-making in unjustified causal claims can lead to severe misunderstandings (Maziarz 2017b).
To address the questions (1) what methods are employed by economists for causal inferences? (2) which of the philosophical theories of causality is preferred by economists? moreover, (3) what philosophical presuppositions underlie the methods of causal inference employed by economists, I conducted the following research. First, from the sample consisting of articles published by American Economic Review (AER), Journal of Political Economy (JPE) and Quarterly Journal of Economics (QJE) since 2005 until 2015, I have chosen those analyzes that conclude causally, i.e., they either use one of the terms labeled by Hoover (2004) "causal family", or use pseudo-causal terms listed by Reiss (2015), i.e., produce, result in, effect, etc. Second, I have analyzed and classified the methods employed by the authors of chosen articles. In line with Hamermesh (2013), the dominance of empirical research in the ways of causal inference is visible. Fourth, I have reconstructed the philosophical presuppositions underlying the methods of causal inference and attempted at indicating which of the philosophical theories of causality is acknowledged by economists.
Third, I have operationalized main philosophical theories of causality in the context of economic research. There are six main theoretical approaches (families of theories) to causality in philosophy that are relevant for the context of economics: regularity theories, probabilistic theories, counterfactual theories, mechanistic theories, interventionist and manipulability theories, and power/capacity approach. Each method of causal inference justifies a specific approach to causal inferences and is connected to different philosophical presuppositions on causality and economic reality. For instance, the Granger-causality tests (Granger 1969; 1980) employ the definition of causality promoted by the probabilistic account (e.g., Sims 1972), cf. Maziarz (2015). Additionally, the project of theoretical macroeconometrics (for instance: the reduced-form vector autoregression models) is also grounded in this approach (cf. Cooley and LeRoy 1985). In contrary to these recent developments in the theoretical econometrics, the usual Cowles Commission approach is grounded in the regularity approach to causality that dates back to the Humean stance. The Cowles Commission approach was developed by Simon (1977) who employed the manipulationist approach to causality and argued in favor of using interventions to solve the indeterminacy of causal direction by data.
The primary results are as follows. First, the mainstream economists draw causal conclusions mainly from observational data. To do so, they usually employ econometric techniques (atheoretical econometrics) that are grounded in the probabilistic approach to causality. A new trend is applying the Bayesian networks (DAG) methods that hitherto was identified with other social sciences. Additionally, several analyses are grounded in either experimental or quasi-experimental methods or case studies. They employ Woodward's (2005) interventionist account.

Bibliography
Claveau, F., & Mireles-Flores, L. (2014). On the meaning of causal generalisations in policy-oriented economic research. International Studies in the Philosophy of Science, 28(4), 397-416.
Cooley, T. F., & LeRoy, S. F. (1985). Atheoretical macroeconometrics: a critique. Journal of Monetary Economics, 16(3), 283-308.
Granger, C. W. J. (1980). Testing for causality. A personal viewpoint. Journal of Economic Dynamic and Control, 2(4), 329-352.
Granger, C. W. J. (1969). Investigating causal relations by econometric models and cross-spectral methods. Econometrica, 37(3), 424-438.
Hamermesh, D. S. (2013). Six decades of top economics publishing: Who and how?. Journal of Economic Literature, 51(1), 162-172.
Hoover, K. D. (2004). Lost causes. Journal of the History of Economic Thought, 26(2), 149-164.
Hutchison T. (2000). On the Methodology of Economics and the Formalist Revolution. London: Edward Elgar Publishing.
Maziarz, M. (2015). A review of the Granger-causality fallacy. The journal of philosophical economics: Reflections on economic and social issues, 8(2), 86-105.
Maziarz, M. (2017a). Causation in Economics. The Most Recent Analyses and the Unsolved Problems. Annales. Ethics in Economic Life, 20(1).
Maziarz, M. (2017b). The Reinhart-Rogoff controversy as an instance of the "emerging contrary result" phenomenon. Journal of Economic Methodology, 24(3), 213-225.
Reiss, J. (2015). Causation, evidence, and inference. London: Routledge.
Sims, Ch. (1972). Money, income, and causality. The American Economic Review. 62(4), 540-552.
Verreault-Julien, P. (2017). Non-causal understanding with economic models: the case of general equilibrium. Journal of Economic Methodology, 1-21.
Woodward, James. Making things happen: A theory of causal explanation. Oxford university press, 2005.

Chair: Till Gallasch
Time: 19:00-19:30, 12 September 2018 (Wednesday)
Location: SR 1.005

Mariusz Maziarz 
(Wroclaw University of Economics, Poland)

I am a philosopher interested in the philosophy of economics (as a principal investigator of a research grant "Causal inferences in the contemporary economics") and an economist (Ph.D. student) interested in the methodology of the discipline.

Testability and Meaning deco