For most contemporary economists, financial instability is an exceptional event that results from exogenous shocks, market imperfections, and/or price instability. Thus, in order to have financial stability, we should promote market mechanisms and correct imperfections, and should have a central bank that guarantees price stability. This position, however, ignores the important contributions of Hyman P. Minsky, John Kenneth Galbraith, John Maynard Keynes and other authors. They show that there are deep causes of financial instability rooted in psychological, sociological, and political forces, which effects are multiplied by the way the capitalist economic system works.
This paper critically analyzes the imperfection view of financial stability by focusing on its analysis of individual behaviors and of the treatment of information. In addition, an alternative way to analyze and to solve problems related to financial stability, the uncertainty view, is presented. The first part of the paper briefly reviews the views on financial stability in the imperfection view. The second part presents some of the limitations of the imperfection view and details some of the key points of the uncertainty view. The third part extracts from the uncertainty view some policy recommendations to promote financial stability.
Market Imperfection as the Source of Instability
In a pure and perfect competitive setting, a "well-behaved" economic system is at equilibrium now and forever. Indeed, economic agents are assumed to make "informed" decisions based on "fundamentals" provided by the economic, political, and financial outlooks. The latter are strong a priori attractors that nobody can ignore without experiencing harsh consequences in a more or less immediate future (and the more efficient a market, the sooner harsh consequences materialize). Thus, in a pure and perfect competitive setting, "both market participants and supervisors will see hints of errant investment strategies of financial problems as they begin" (Shinasi 2006, 179) and so "market discipline" will do its job to eliminate any unsustainable decisions, that is decisions not in accordance with the fundamentals.
Behind this reasoning lie specific assumptions about the nature of human rationality and the nature of the information provided to economic agents. First, following methodological individualism, economic agents are assumed to be robot-like beings that respond to price signals according to their own preference, without any considerations for their social environment. It is possible to partly deviate from this assumption by having economic agents that follow Bayesian probabilities (as the "cascade of information" literature does), or by taking into account some of the "cognitive biases" or "anomalies" observed (as behavioral finance does). However, over time the bounded rationality of individuals with cognitive biases is corrected by additional information. Second, the information received is a neutral input directly usable to feed the decision process in order to generate an output (buy, sell, or stay neutral). Third, changes in prices are assumed to give a direct understanding of the underlying motivations of economic agents.
In this context, it is very difficult to explain financial instability as a process and to understand why there are recurring periods of growing financial fragility followed by financial instability. financial instability is an exogenous state created by the bounded rationality of economic agents and the imperfection of information. Indeed, without the capacity to input information and/or "without sufficient timely information, the market disciplining mechanisms [...] might not produce the appropriate self-corrective adjustments" (Schinasi 2006, 167), which will lead economic agents to undertake unsustainable decisions and thus prevent the efficient allocation of real resources from savers to investors.
In terms of policy, this implies that financial stability can be improved by allowing market mechanisms to work more fully. This can be done by improving the quality and quantity of information, improving the computing capacity of individuals, and educating economic agents so that their cognitive biases are eliminated (Shiller 2000; Schinasi 2006; Warneryd 2001). Public authorities can help to ensure...