Economic modeling is the foundation in building economic analysis and insights. The reading material represents a spectrum of views on the role of realism and assumptions in economic modeling. This paper explores various perspectives on this issue, highlighting the importance of balancing predictive accuracy with realistic assumptions to develop robust and meaningful economic models.
Mill (1844) started by advocating for deductive reasoning based on simplified assumptions, stating that “a priori is a legitimate mode of philosophical investigation in the moral science”. Later on, Friedman (1953) advocates an instrumentalist view, prioritizes predictive accuracy above all else. He argues that the realism of a model’s assumptions is irrelevant; the only criterion for judging a theory is its predictive power. A model is “good” if it predicts well, regardless of whether its assumptions are a true reflection of reality.
However, I disagree with Friedman’s (1953) argument that the realism of assumptions in economic models is irrelevant. While predictive accuracy is important, dismissing the realism of assumptions entirely can lead to wrong conclusions. Understanding causal relationships between assumptions and prediction result is very important in economics, not only its correlations. For example, a model might mistakenly predict the rice-agriculture production based on a pine trees height growth in the same region. When the pine trees height grow by 5% per year, the rice production grows by a similar growth. Although there is a correlation between the pine trees and rice field (both effected by weather), it may not be necessarily true that the pine trees causes the increase in rice agricultural production. A better way to predict is by directly analysing the causal relationship between the weather and rice production.
Furthermore, I largely agree with Hausman’s (2008) arguments in Why Look Under the Hood. He argues that understanding the underlying mechanisms and assumptions is important, even if a model predicts well. He emphasizes the need to understand causality, evaluate robustness, improve model building, and avoid misapplication. This is especially relevant in today’s economics, as variables moves so fast in complex direction. Only by understanding the model in detail, we can predict correctly. For example, a model predicting stock prices need to incorporate behavioural factors and market imperfections, rather than relying solely on the efficient market hypothesis. When we do not include the right assumptions, predictions will behave inconsistently given the complexity of human and market behaviour. It is actually important to have realists assumption when making predictions.
Lastly, Maki (2009) did an excellent job in rewriting Friedman’s (1953) essay, in a way that is consistent with realist interpretation. I strongly agree with Maki (2009) on his critique of “blackboard economics”, “a mathematically highly refined and rigorous but accused of being unconnected to real-world facts and issues”. For example, the idea of perfect rationality and utility maximization in economics is simple in theory, yet very different in real life. The theory imagines a world where everyone is a perfect robot shopper. These robot shoppers always make the absolute best decision possible, maximizing their utility (which is basically their happiness or satisfaction) with every purchase. They have perfect information about every product, know exactly what they want, and flawlessly calculate the best way to get it. Yet, we know this is far from how humans operate.
In conclusion, the diverse perspectives on the role of realism and assumptions in economic modeling contribute significantly to the evolution of economic thought. These discussions underscore the need for a balanced approach that values both predictive power and the realism of assumptions. By integrating these perspectives, economists can develop models that not only predict accurately but also provide meaningful insights into real-world phenomena.
Reference
Mill, J.S. (1844). “On the Definition of Political Economy and the Method of Investigation Proper to It,” in Essays on Some Unsettled Questions of Political Economy.
Friedman, M. (1953). “The Methodology of Positive Economics,” in Friedman, Essays in Positive Economics. Chicago: University of Chicago Press.
Hausman, D. (2008). “Why Look Under the Hood?” in Essays on Philosophy and Economic Methodology. Cambridge: Cambridge University Press, ch. 5 (pp. 70-74). Uskali Mäki,
Maki, U. (2009). “Unrealistic Assumptions and Unnecessary Confusions: Rereading and Rewriting F53 as a Realist Statement,” in Uskali Mäki, editor, The Methodology of Positive Economics: Reflections on the Milton Friedman Legacy. Cambridge: Cambridge University Press, ch. 3 (pp. 90-116).
Leave a Reply