INET Announces Program on Knightian Uncertainity Economics

Rethinking the role of markets and government policy in light of our inherently limited ability to foresee economic and social outcomes

The new INET Program on Knightian Uncertainty Economics (KUE) is inspired by arguments advanced by Frank Knight, John Maynard Keynes, Friedrich Hayek, and Karl Popper about the inherent limits of what we can know about the future. As Popper argued, “Quite apart from the fact that we do not know the future, the future is objectively not fixed. The future is open: objectively open.”

The Program focuses on the key implication of this openness for macroeconomics and finance theory: The processes driving aggregate outcomes undergo change at times and in ways that cannot be characterized as a standard (probabilistic) risk. Consequently, the Program aims to develop formal macroeconomics and finance models and approaches to policy analysis that recognize that economists, policymakers, market participants, and citizens more broadly face “true”—Knightian—uncertainty arising from such unforeseeable change. By advancing this research, the INET Program on KUE seeks to contribute to a much-needed reconsideration of the role of markets and government policy in modern capitalist economies.

The Program on KUE has evolved from the previous phase of INET-supported research, which aimed to develop an approach—called Imperfect Knowledge Economics (IKE)—that would place unforeseeable change, and the Knightian uncertainty that such change engenders, at the center of macroeconomic and policy analysis. The research produced by the Program on IKE focused primarily on a critique of the milestone approaches—the rational expectations hypothesis and behavioral finance—to macroeconomic and finance theory since the 1970s. Although these approaches differ in a number of essential respects, they share a common feature: Their models represent outcomes with a stochastic process, thereby assuming away Knightian uncertainty.

A number of arguments have been advanced to show that prevailing macroeconomic models suffer from epistemological flaws that render their ability to explain how outcomes actually unfolded in the past—let alone to predict future values—limited at best. The economics profession has responded by focusing on arguably important missing features of the prevailing models, such as the near absence of the financial sector in the widely-used Dynamic Stochastic General Equilibrium models.

However, economists have been reluctant to consider the possibility that probabilistic representations of uncertainty, which underpin all of the prevailing models, miss something more essential: that the process driving outcomes undergoes change at times and in ways that cannot be characterized with a stochastic process. In contrast, IKE traced the theoretical and empirical difficulties of the prevailing models to their core premise that outcomes do not undergo unforeseeable change and thus that market participants, economists, and policymakers do not face “true”—Knightian—uncertainty. Roman Frydman and Michael D. Goldberg presented the theoretical and empirical analysis of these difficulties in two books: Imperfect Knowledge Economics: Exchange Rates and Risk, and Beyond Mechanical Markets: Asset Price Swings, Risk, and the Role of the State, which were published by Princeton University Press in 2007 and 2011, as well as in a number of papers.

The arguments advanced by IKE’s critique have played an important role in developing an approach that would enable economists and policymakers to build models that recognize that they face Knightian uncertainty. After a multiyear effort, Roman Frydman, Søren Johansen, Anders Rahbek, and Morten Tabor have completed the development of such an approach—called the Knightian Uncertainty Hypothesis (KUH). Hence, INET’s research program to develop an alternative to the prevailing paradigm in macroeconomics and finance theory is being relaunched as the INET Program on KUE.

Please also see INET President Rob Johnson’s introductory commentary regarding the program, “Why We Need the Knightian Uncertainty Hypothesis.”