Microsoft Research New England; University of Chicago
July 28, 2015
I survey the literature on “price theory,” defined as neoclassical microeconomic analysis that reduces rich and often incompletely-specied models into “prices” sufficient to characterize approximate solutions to simple allocative problems. I argue that this definition is more descriptively accurate today, coherent through history and useful than are alternative definitions. I illustrate my definition by showing how it naturally leads to price theory’s distinctive and familiar characteristics: its affinities to other disciplines, approach to measurement and use of diagrams and problem sets. I trace the origins of price theory from the early nineteenth century through its segregation into the Chicago School in the last quarter of the twentieth. I suggest that this relative scarcity combined with natural complementarities between price theory and work outside the tradition have fueled a recent resurgence of price theory in fields ranging from market design to international trade. This, in turn, has raised the value of developing more rigorously the approximation methods used in price theory, a literature still in its infancy.
Price Theory – Introduction
The whole state of the universe at one moment may perhaps be said to cause its whole state at the next moment, but when we say “A” is the “cause” of “B” we always assume that other things are equal…”The” cause of a phenomenon is merely that one of its necessary conditions which is for some practical reason crucial, generally from the standpoint of control.
– Frank Knight, Risk, Uncertainty, and Profit
This project is dedicated to the memory of Gary S. Becker, who first inspired my interest in price theory.
In this paper I survey the contributions and methods of the literature on “price theory”. This literature has deep historical roots, tracing back to the early nineteenth century, and has recently seem a resurgence across a broad range of fields of economics, from international trade to market design, especially through the work of five of the seven most recent winners of the John Bates Clark Medal. Yet despite its common usage, appearing according to Google’s NGrams viewer twice as frequently as “labor economics”, I am not aware of any broadly accepted definition of the phrase. I therefore offer my own working definition, which this survey will employ, explain and illustrate: analysis that attempts to simplify a rich (high-dimensional heterogeneity, many agent, dynamics, etc.) and often incompletely specified model for the purposes of answering a simple (scalar or unidimensional) allocative question.
Even when they are not denominated in currency, these summaries function as “prices” in the sense of von Mises (1920) and Hayek (1945): they provide parsimonious means of communicating rich economic data for limited allocative purposes. Deriving such prices contrasts both with work that richly characterizes complete but often simple economic models and work that takes induction from empirical measurement as prior to the theory.
This definition tries to be simultaneously as consistent as possible with contemporary usage, as coherent as possible across history and as useful as possible in suggesting a research agenda. While there were trade-offs among these criteria, along all of these three dimensions, I believe my definition is both relatively more satisfactory than existing attempts at a definition and performs well in absolute terms.
To see this consider an alternative perspective that has recently gained some currency in the literature on the history of economic thought, emphasizing the price-taking, partial equilibrium analysis that was emphasized at the University of Chicago in the 1950’s and 1960’s (Grossbard, 2010; Medema, 2011; Hammond et al., 2013). While an accurate description of most Chicago price theory in the 1950’s and 1960’s this definition does not fit well with contemporary usage, in which, for example, the work of Jeremy Bulow on imperfect competition and mechanism design is closely associate with price theory. By contrast as I illustrate in Section III, my definition ties together work from a wide range of substantive fields that is identified by contemporary economists as having a price theoretic flavor.
Nor was there any time period in which any major figure identified the phrase “price theory” per se with price taking and partial equilibrium analysis. Even in the relevant period, Milton Friedman (Hammond, 1992) argued that these distinctive features were part of Chicago Price Theory, but not price theory more broadly. By contrast as I highlight in the next section, my definition coheres with most work that has been identified as price theoretic widely within the profession across most history since the early nineteenth century.
Finally, the standard definition limits price theory to topics and approaches that no longer constitute an important focus of most economic research. Thus they are mostly of historical interest. By contrast, my definition helps make sense of the complementarity of price theory with advances in other methodologies of economic analysis that has helped fuel a resurgence in price theoretic work the last decade. Perhaps more importantly, it also highlights the methodological tools that are central to price theory. In Section IV below I argue that lack of clarity on the role of these methods has led them to be underdeveloped, suggesting directions for future research.
A useful way to understand my definition is through several manifestations of price theory’s characteristic drive to obtain simple, approximate answers to simple questions in rich settings. I summarize these properties in Table 1, which contrasts these characteristics to attitudes towards these questions in other neoclassical microeconomic analysis. I now discuss these using illustrations from both historical and recent work in the tradition. My focus throughout this piece is exclusively on work that would be classified by most as microeconomics, as a range of different methodological traditions exist in macroeconomics on which I am far from being an expert and which cut in different directions from the distinction I am trying to draw.
See full PDF below.