Opening Models of Asset Prices and Risk to Non-Routine Change


Paper revised for the Institute’s Plenary Conference in Berlin

Financial markets are, in most respects, prototypes of the markets for which much of contemporary economic analysis was designed. They are characterized by a large number of buyers and sellers; powerful monetary incentives; few, if any, barriers to entry and exit; no impediments to the adjustment of prices; and a plethora of available information that is quickly disseminated around the world. We would expect that financial markets would offer the best opportunity for contemporary economic models to provide explanations of market outcomes. But it is precisely in these markets that contemporary macroeconomic and finance theory have encountered many of their most glaring empirical difficulties. In our companion paper for this conference, we trace contemporary economic theory’s empirical and epistemological problems to how it models change in the causal process underpinning market outcomes.1 In fi- nancial markets, outcomes are driven primarily by market participants’ forecasts of prices and risk. As time passes, participants revise their forecasting strategies in ways that they themselves, let alone an economist, cannot fully foresee. Economic policies, institutions, the state of technology, and other features of the social context within which participants make decisions also change in novel ways. Thus, change in financial markets, and in capitalist economies more broadly, is, to a significant degree, non-routine, for it cannot be adequately represented in advance with mechanical rules and procedures. Yet, the hallmark of contemporary theory is the core premise that an economist can fully specify, in terms of some causal factors, how individuals alter the way that they make decisions, and how market outcomes unfold over time. In this paper, we follow an alternative approach to economics analysis — called Imperfect Knowledge Economics (IKE) — and develop a model of asset prices and risk that has explicit mathematical microfoundations, and yet remains open to non-routine change.

Share your perspective