Economists have struggled to characterize and model the dynamic evolution of economic phenomena throughout this century. From the earliest attempts to establish business annals at the beginning of the century to the disappointing appraisals of recent empirical research on efficient markets for financial assets (involving rational expectations and bubbles) circa 1990, the numerous representations of evolving prices, for example, have been forced to acknowledge the hazards of systemic instability, excessive volatility, and nonstationarity. As well, a host of technical issues stem from the weak numerical foundations for all econometric specifications of market fundamentals; the implicit dynamics of invisible-hand mechanisms, the inevitable reliance on flawed and incomplete data, the adverse consequences of both temporal and spatial aggregation over market participants with heterogeneous and vague information, and potential fragility of financial markets. For at least eight decades, American economists have faced the persistent choice between structural or formal models of evolving dynamics (the explorations of which have often been determined by a strong reliance on correlation, regression, and their later probabilistic counterparts) and those alternative portrayals that focused on historical narratives or qualitative features. These latter approaches seem better suited for the treatment of financial convulsions, crises, and manias, but may preclude the derivation of strong generalizations from past circumstances.
In the following comments, we restrict our attention, in the main, to some representative authors with high profiles who have sought to address the swings in stock prices. Thus, we omit most discussion of other important efforts to deal with wider issues reflecting long historical perspectives, the comparative search for "explanatory power," and contemporary literature on banking crises. For example, we exclude Kindleberger (1989), who adopted Minsky's framework in seeking to explain major historical crises over a period of more that two centuries; Wolfson (1986), who compared the "success" of various theories in describing most postwar American crises; much of the orthodox empirical work illustrated by LeRoy (1989) and West (1988); the interest in bank runs of Diamond and Dybvig (1983); and the familiar approach to American financial crises of Friedman and Schwartz (1963). Despite such omissions, we are left with a considerable range for alternative visions of evolving markets, diverse methodological approaches, and attitudes to empirical findings with which to describe the diverse character of some economic theories for stock prices. Our particular choices reflect both a short-term reaction to a widening recognition of the empirical failures for most fashionable combinations of efficient markets and rational expectations in relation to stock-price volatility, and longer-term personal interests in the nonquantitative literature on manias, crises, and panics that began with Jain (1985) and Spotton (1993).
In the first two sections, we identify some of the existing theories and group them into broad "orthodox" and "heterodox" categories, before comparing and contrasting the major attributes that characterize these two groups. Then, in the third and fourth sections, respectively, we establish the principal patterns of financial instability and provide a tentative appraisal. A few terse remarks conclude.
Some Existing Theories
The causes of manias and crashes in stock markets, and of systemwide bank failures and other symptoms of financial instability remain imperfectly understood. Our current failures to fully understand such phenomena do not result from neglect in the sense that economists have been remiss in ignoring many of the relevant issues here. Active interest in them can be traced back at least to the early nineteenth century - as illustrated by Mackay (1852), Burton (1902), Sprague (1910), Cowles (1933), Fisher (1930, 1932, 1933), and Schumpeter (1931), for example - and continues, unabated, to the present. …