Rational expectations

The stampede toward "rational expectations" — widely thought to be a "revolution," though it was only a generalization of the neoclassical idea of equilibrium—derailed the expectations-driven model building that had just left the station. In the end, this way of modeling has not illuminated how the world economy works.
- Edmund S. Phelps (2007)

Rational expectations is a hypothesis which states that agents' predictions of the future value of economically relevant variables are not systematically wrong in that all errors are random.

CONTENT : A–F , G–L , M–R , S–Z , See also , External links

Quotes

Quotes are arranged alphabetically by author.

A–F

• One troublesome aspect is the place of rational expectations macroeconomics in the often political debate over Keynesian economics. At least implicitly, many people feel that what's bad for the rational expectations viewpoint is good for the Keynesian one, and vice versa. But it is hard to see how the problems in using the rational expectations approach to explain monetary nonneutrality can alleviate the theoretical and empirical shortcomings of the Keynesian model.
  • Robert J. Barro, "Rational Expectations and Macroeconomics in 1984" (1984).

• Using the popular macroeconomic models of the time, Lucas and Sargent showed how replacing traditional assumptions about expectations formation by the assumption of rational expectations could fundamentally alter the results. … Most macroeconomists today use rational expectations as a working assumption in their models and analyses of policy. This is not because they believe that people always have rational expectations. Surely there are times when people, firms, or financial market participants lose sight of reality and become too optimistic or too pessimistic. … But these are more the exception than the rule, and it is not clear that economists can say much about those times anyway. When thinking about the likely effects of a particular economic policy, the best assumption to make seems to be that financial markets, people, and firms will do the best they can to work out the implications of that policy. Designing a policy on the assumption that people will make systematic mistakes in responding to it is unwise.
  • Olivier Blanchard, Macroeconomics (7th Edition, 2017), Ch. 16 : Expectations, Output, and Policy

• So monetary policy decisions tend to regress toward the mean and to be inertial—and hence biased in just the same way that adaptive expectations are biased relative to rational expectations. But errors like that, while systematic, will generally be small and will tend to shrink over time. And, in return, the system builds in natural safeguards against truly horrendous mistakes.
  • Alan S. Blinder, Central Banking in Theory and Practice. 1998

• Most mainstream macroeconomic theoretical innovations since the 1970s (the New Classical rational expectations revolution associated with such names as Robert E. Lucas Jr., Edward Prescott, Thomas Sargent, Robert Barro etc, and the New Keynesian theorizing of Michael Woodford and many others) have turned out to be self-referential, inward-looking distractions at best. Research tended to be motivated by the internal logic, intellectual sunk capital and esthetic puzzles of established research programmes  rather than by a powerful desire to understand how the economy works – let alone how the economy works during times of stress and financial instability.  So the economics profession was caught unprepared when the crisis struck.
  • Willem Buiter, "The unfortunate uselessness of most ‘state of the art’ academic monetary economics" (2009)

• Its development of Mises’s theory of action anticipated much of the rational expectations idea that would later win Robert Lucas a Nobel Prize.
  • Eamonn Butler, Austrian Economics: A Primer.

• The current macroeconometrics literature has proposed two ways to confront general equilibrium rational expectations models with data. The first, an estimation approach, is the direct descendant of the econometric methodology proposed 50 years ago by Haavelmo (1944). The second, a calibration approach, finds its justification in the work of Frisch (1933) and is closely linked to the computable general equilibrium literature surveyed e.g. in Shoven and Whalley (1984).
  • Fabio Canova, "Statistical inference in calibrated models." Journal of Applied Econometrics 9.S1 (1994): S123-S144.

• The reason is that the second order properties induced by similar general equilibrium models are, in general, so different from those of actual data that some authors (e.g. Campbell and Shiller 1987, Frankel and Froot 1987) have concluded that the simple version of the rational expectations-efficient market hypothesis is severely flawed.
  • Fabio Canova. "Sensitivity analysis and model evaluation in simulated dynamic general equilibrium economies." International Economic Review (1995): 477-501.

• The Keynesian econometric methodology developed by Klein and associates was criticized by Lucas in his 1976 “Critique of Econometric Policy Evaluation” on the grounds that microfounded structural equations should contain expectations of future variables. Since the parameters of these expectations should depend on the parameters of the rules followed by the policy authorities, Lucas argued that the rational expectations assumption would invalidate the practice of using fixed parameter models as policy guides.
  The profession responded to the Lucas critique in two different ways. The first, introduced to economists in the book Rational Expectations and Econometric Practice, was to develop appropriate econometric methods to estimate parameters in rational expectations environments. The second, explained most clearly in Kydland and Prescott’s (1996) article, “The Computational Experiment,” was to develop a new methodology,calibration, that lowered the standards of what it means for a model to be successful by requiring that a good model should explain only a limited set of empirical moments.
  • Roger Farmer, Expectations, Employment and Prices, Ch. 5 A New Way to Understand Business Cycle Facts.

• The rational expectations revolution of the 1970s threw out the textbook Keynesian apparatus because it did not cope well with the simultaneous appearance of high inflation and high unemployment in the 1970s. But the model could not have been rejected so quickly on empirical grounds if it was not already on weak theoretical foundations.
  • Roger Farmer, Expectations, Employment and Prices, Ch. 6 The Great Depression: Telling the Keynesian Story in a New Way.

• In the early 1970s,... economists began to embrace the Rational Expectations Hypothesis (REH), according to which market participants’ expectations are “essentially the same as the predictions of the relevant economic theory” (Muth 1961: 316). What has been largely overlooked is that, … REH theorists presume that the role of market participants’ expectations in driving outcomes is not autonomous from the other components of the model. … Because REH models, by design, rule out an autonomous role for expectations, they are best viewed as derailing, rather than developing, the microfoundations approach.
  • Roman Frydman and Edmund S. Phelps, "Which Way Forward for Macroeconomics and Policy Analysis?" in Rethinking Expectations: The Way Forward for Macroeconomics edited by Roman Frydman and Edmund S. Phelps.

G–L

• Nonetheless, as Paul Krugman’s popular writings have indicated,economists’ confidence in their ability to guide the economy and to advance the commonweal had significantly weakened after Heller’s 1965 Godkin Lectures, mentioned earlier. The discovery of the “natural rate of unemployment” and development of the theory of rational expectations revealed the limitations of economists’ macroeconomic policy tools.[The doctrine of rational expectations posits that the market will always anticipate government policy and will neutralize its intended effects.] Moreover, Krugman bemoaned the fact that “policy entrepreneurs” frequently displaced economists in providing economic advice to society. Referring to supply-side economics and other questionable economic doctrines, Krugman, using less than elegant words, suggested that a major task for economists must be “to flush such economic cockroaches down the toilet.
  • Robert Gilpin, Global political economy : understanding the international economic order, Ch 3. The Neoclassical Conception of the Economy.

• The attack on Keynesian economics in the 1960s and 1970s by monetarists and by the theory of rational expectations undermined the intellectual foundations of development economics. The essence of this criticism was that there is only one economics, and that economics is a universal science equally applicable to all societies.
  • Robert Gilpin, Global political economy : understanding the international economic order, Ch. 12. The State and Economic Development.

• The original Lucas version of the new-classical macroeconomics combined the undeniable appeal of rational expectations with two more dubious assumptions inherited from Friedman (1968), that is, continuous market clearing and imperfect information, to form the foundation of the famous “Lucas supply function” (more justly, the Friedman-Lucas supply function). Soon Sargent and Wallace (1975) extracted from Lucas’s model its implication for monetary policy, the famous “policy-ineffectiveness proposition.” The demonstration by Barro (1977) that one could interpret historical U.S. data to be consistent with the proposition and the theory brought new-classical economics to its shortlived period of peak influence.
  • Robert J. Gordon, “Fresh Water, Salt Water, and other Macroeconomic Elixirs.”Economic Record. March 1989; pp. 177–84.

• A contrast with John Taylor’s approach to investigating the properties of models is instructive. Taylor’s research program includes the use of computational experiments. It is well summarized in his recent book (Taylor, 1993). Like Kydland and Prescott, Taylor relies on fully specified dynamic models and imposes rational expectations when computing stochastic equilibria. However, in fitting linear models he uses all of the information on first and second moments available in the macro data when it is computationally possible to do so. The econometric methods used in parameter estimation are precisely described. Multiple sources of business cycle shocks are admitted into the model at the outset, and rigorous empirical testing of models appears throughout his analyses.
  • Lars Peter Hansen and James J. Heckman. "The empirical foundations of calibration." The Journal of Economic Perspectives (1996): 87-104.

• Friedman’s monetarist model is distinguished from Lucas’s new classical monetary model in that Friedman imagines that people can be systematically mistaken about the true state of real wages for relatively long periods, while Lucas argues that people have rational expectations (i.e., they make only unsystematic mistakes) and, therefore, correct their judgments about the real wage quickly.
  • James Hartley, Real Business Cycles: A Reader. Chapter 1. The limits of business cycle research

• Unlike in the Solow model, the factors important to the savings decision now enter separately from those important to the investment decision. Aggregate demand pathologies are, nonetheless, impossible, because in Lucas’s model the same agents make both the savings and the investment decision, which insures ex ante coordination, and the agents have rational expectations, which insures that mistakes about the future course of the economy are necessarily unsystematic. Furthermore, the supply of labor responds elastically to temporarily high real wages: workers make hay while the sun shines.
  • James Hartley, Real Business Cycles: A Reader. Chapter 1. The limits of business cycle research

• Households’ decisions are more complicated: given their initial capital stock, agents determine how much labor to supply and how much consumption and investment to purchase. These choices are made in order to maximize the expected value of lifetime utility. Households must forecast the future path of wages and the rental rate of capital. It is assumed that these forecasts are made rationally. A rational expectations equilibrium consists of sequences for consumption, capital, labor, output, wages, and the rental rate of capital such that factor and output markets clear.
  • James Hartley, Real Business Cycles: A Reader. Chapter 2. A user’s guide to solving real business cycle models.

• I could tell you a story about my friend and colleague Milton Friedman. In the nineteen-seventies, we were sitting in the Ph.D. oral examination of a Chicago economist who has gone on to make his mark in the world. His thesis was on rational expectations. After he’d left, Friedman turned to me and said, “Look, I think it is a good idea, but these guys have taken it way too far.”

  It became a kind of tautology that had enormously powerful policy implications, in theory. But the fact is, it didn’t have any empirical content. When Tom Sargent, Lard Hansen, and others tried to test it using cross equation restrictions, and so on, the data rejected the theories. There were a certain section of people that really got carried away. It became quite stifling.

  • James Heckman, in "Interview with James Heckman" by John Cassidy (January 14, 2010).

• In models of the use of information by rational agents, it is generally assumed that all will interpret the same signals in the same way. In the extreme case of the ‘rational expectations hypothesis’, it is held that through mere data-gathering, agents will become aware of the basic, underlying structure and mechanisms of the economy. This hypothesis likewise neglects the conceptual framing involved in the perception of data and the theory-bound character of all observation.
  • Geoffrey M. Hodgson, "The evolution of capitalism from the perspective of institutional and evolutionary economics" in Capitalism in evolution: Global contensions–East and West (2001).

• The whole of academic macroeconomics is touched by the rational expectations hypothesis. Althought it has been criticized on conceptual grounds and as emprically inadequate, the rational expectations hypothesis sets a noncontroversial standard for modeling expectations in macroeconomics.
  • Kevin D. Hoover, "The Rational Expectations Revolution: An Assessment" (1992).

• Muth is the Rosseau of the rational expectations revolution; Lucas is its Robespierre.
  • Kevin D. Hoover, "The Rational Expectations Revolution: An Assessment" (1992).

• Along with the shift in focus to investigating the sources and nature of business cycles, aggregate analysis underwent a methodological revolution. Previously, empirical knowledge had been organized in the form of equations, as was also the case for the early rational expectations models. Muth (1960), in his pioneering work on rational expectations, did not break with this system-of-equations tradition. For that reason, his econometric program did not come to dominate. Instead, the program which has prevailed is the one that organizes empirical knowledge around preferences, technology, information structure, and policy rules or arrangements. Sargent (1981) has led the development of tools for inferring values of parameters characterizing these elements, given the behavior of the aggregate time series. As a result, aggregate economics is no longer a separate and entirely different field from the rest of economics; it now uses the same tools and empirical knowledge as other branches of economics, such as finance, growth theory, public finance, and international economics. With this development, measurements and quantitative findings in those other fields can be used to restrict models of business cycles and make our knowledge about the quantitative importance of cyclical disturbances more precise.
  • Finn E. Kydland and Edward C. Prescott, "Business cycles: Real facts and a monetary myth." (1990).

• Under perfect foresight (rational expectations), a regime of sequential loan markets and spot markets in labor services also supports the optimal solution as a competitive equilibrium.
  • Robert G. King, Charles I. Plosser, and Sergio T. Rebelo. "Production, growth and business cycles: I. The basic neoclassical model." Journal of monetary Economics 21.2 (1988): 195-232.

• At this point, readers may want to take a deep breath. Lucas's conclusions are far more sweeping than anything Friedman proposed, and his chain of logic seems much longer. Is it necessarily right?
  The short answer is, of course, no. There are at least two weak links in the chain. The first is the proposition that a recession lasts only as long as firms are confused about the actual economic situation—in effect, that you can have a recession only as long as most people don't notice. This idea was implicit in the way that Friedman presented his natural rate argument, but it is something that few people other than economists find plausible (and it is possible to accept Friedman's views about stagflation without fully buying his notion that monetary policy works only by fooling people). The second weak link is the whole idea that firms set their prices by closely watching monetary policy or, worse yet, macroeconomic indicators that might help predict monetary policy.
  • Paul Krugman, Peddling Prosperity (1994), Ch. 1 : The Attack on Keynes

• In the late 1970s Lucas and his disciples worked feverishly, attempting to produce theoretical models in which firms and households can observe stock prices and interest rates, yet still be subject of the kind of rational confusion needed to create booms and slumps. They failed. The theory underlying a rational expectations business cycle just didn't work out.
  At the same time, it was becoming increasingly obvious that the rational expectation story didn't work in practice either. Here the difficulty is easier to explain: economic slumps last too long.
  • Paul Krugman, Peddling Prosperity (1994), Ch. 8 : In the Long Run Keynes is Still Alive

• The hypothesis was more or less buried during the '60s. Arrow used it in his paper on learning-by-doing in the '60s. Prescott and I used it in that paper of ours on investment. People were aware of it, I didn't understand then how fundamental a difference it mad econometrically. I didn't realize that if you took it seriously you had to rethink the whole question of testing and estimation. I guess no one else did either, except for Muth.
  • Robert E. Lucas, in Conversations with Economists (1983) by [[Arjo Klamer]

• Leonard Rapping’s and my contribution to that volume was a partial equilibrium model of the labor market only. The idea was that agents had to base production and employment decisions on partial information and that this could lead them to decisions that to an omniscient observer would appear pathological. The new elements in my paper, relative to Phelps’s outline, were its mathematical explicitness and its use of rational expectations. The example developed in the paper made it clear that the possibility of using monetary policy to stimulate production would depend on the information people have and could not be reliably described by a stable Phillips curve. But attempts to construct quantitative models of recessions based on this feature alone have not succeeded.
  • Robert E. Lucas Jr. Introduction in Collected Papers on Monetary Theory.

M–R

• The "rational expectations" school holds that agents do not know the future, but they formulate their expectations on the basis of a satisfactory knowledge (i.e., a theory) of how the economy functions. If we add propositions to the effect that each agent's decisions are based on its maintained theory of system behavior, that general equilibrium theory is an apt representation of the world, and that those whose behavior is consistent with this apt theory will be successful, then an equilibrium and equilibrating view of the economy emerges. … If the economy does not conform to the general equilibrium theory, if it is endogenously unstable, and if units behave accordingly, then rational expectations will exacerbate instability.
  • Hyman P. Minsky, Stabilizing the Unstable Economy.

• Macro rational expectations, as I have labeled the hypothesis, seems to say that expectations in an economist's model must be perfectly consistent with his model that embodies these expectations. In other words, the agents of his model must all share his views of the relevant economic mechanisms, as well as his data. Why? Because if he holds them they must believe they are God's truth and, if so, rational people can have no other views (and of course we should never ask how they would come by these views and data, that not even other specialists may have heard of yet, let alone accepted). I submit that this view is pretty absurd--I would almost say offensive! I certainly believe that I know more about economics and the economy than (almost) everybody else, and i can even prove it: If everybody shared my vies, then the economy could not be in today's troubles (though it might conceivably be in some different ones!).
  • Franco Modigliani, in Conversations with Economists (1983) by Arjo Klamer

• Interestingly, recent research shows that herd behavior is consistent with rational expectations when information is imperfect, though the extent of the herd behavior may well be greater than can be explained by these models. The essential reason for volatility in financial markets, as emphasized by Keynes, is that market players respond to expectations. The value of any asset today depends on what others are expected to be willing to pay for it tomorrow, and that depends in turn (in a never ending chain) on what others are expected to be willing to pay the day after. These expectations are based on information about current conditions. Such information is inherently incomplete and costly to process. This makes it rational for everyone to glean information about the desirability of investing from the opinions and actions of others.
  • José Antonio Ocampo, Shari Spiegel, and Joseph E. Stiglitz, "Capital Market Liberalization and Development" in Capital Market Liberalization and Development.

• Without capital markets extending infinitely far into the future, the economy will not necessarily converge to the long run equilibrium. There are paths which are dynamically consistent with rational expectations in the short run. While herding behavior is often attributed to investor myopia, these results suggest that bubbles may arise so long as investors do not look infinitely far into the future. However, even when investors look infinitely far into the future, it may not be possible for them to predict (on the basis of rational expectations alone) how the economy will evolve, if, for instance, there are multiple paths consistent with rational expectations.
  • José Antonio Ocampo, Shari Spiegel, and Joseph E. Stiglitz, "Capital Market Liberalization and Development" in Capital Market Liberalization and Development.

• The stampede toward “rational expectations”—widely thought to be a “revolution,” though it was only a generalization of the neoclassical idea of equilibrium—derailed the expectations-driven model building that had just left the station. In the end, this way of modeling has not illuminated how the world economy works.
  • Edmund S. Phelps (2007) “Foreword,”in Roman Frydman and Michael D. Goldberg, Imperfect Knowledge Economics: Exchange Rates and Risk.

• A still more extreme thesis is proposed by rational expectations theoreticians, among whom is the American Robert Lucas (b. 1937, Nobel prize in 1995). In a 1972 article, Lucas joined the assumption of markets in continuous equilibrium with that of rational expectations, originally formulated by Muth (1961), according to which ‘expectations [...] are essentially the same as the predictions of the relevant economic theory’. As a consequence, economic agents learn to take account of public intervention in the economy, discounting its effects beforehand. Thus, for instance, deficit public expenditure, that is not financed by a contemporary increase in taxation, adopted by the government to stimulate aggregate demand, is counterbalanced by a reduction in private consumption, decided by private economic agents to put aside the savings with which to pay for the taxes which sooner or later will have to be introduced to pay for the public debt with which public expenditure is financed. In this context, the Phillips curve turns out to be vertical also in the short run: expansionary monetary and fiscal policy interventions may only produce an increase in the rate of inflation, not in the level of employment. … The rational expectations assumption, in the usual context of a one commodity model, also underlies a new theory of the trade cycle, the ‘real cycle theory’. After dominating the scene in the 1980s, in the following decade rational expectations theory gradually lost ground, even if in the theoretical confrontation with representatives of the neoclassical synthesis the shaky nature of its theoretical foundations – the one-commodity model, common to their rivals too – has not been stressed.
  • A. Roncaglia, "The age of fragmentation" in The Wealth of Ideas.

S–Z

• These ideas have implications not only for theoretical and econometric practices but also for the ways in which policymakers and their advisers think about the choices confronting them. In particular, the rational expectations approach directs attention away from particular isolated actions and toward choices among feasible rules of the game, or repeated strategies for choosing policy variables. While Keynesian and monetarists macroeconomic models have been used to try to analyze what the effects of isolated actions would be, it is now clear that the answers they have given have necessarily been bad, if only because such questions are ill-posed.
  • Thomas J. Sargent, "Rational Expectations and the Reconstruction of Macroeconomics" (1980).

• I was reminded of how much I had misjudged the potential the profession would see in the time series rational expectations models. When I, as a graduate student at the Massachusetts Institute of Technology (MIT) around 1970 did some work on the econometrics of rational expectations time series models, I felt rather apologetic about the extreme assumptions in the models. I did not expect others to regard them as anything more than a passing gimmick. Richard Sutch had just written in his MIT doctoral dissertation (1968) an exposition of the coefficient restrictions implied for time series representations of long-term and short-term interest rates, but he never bothered to publish this work. I remember conversations with him and others about rational expectations models, and I did not come away thinking they were the wave of the future.
  • Robert J. Shiller (1984), Review of Rational Expectations and Econometric Practice by Robert E. Lucas, Thomas J. Sargent.

• Neither historical experiences nor rational expectations models provide much guidance as to how market participants’ expectations will respond in any particular situation. Each situation has its own distinctive characteristics, and there is ample evidence that many market participants’ expectations do not conform to the strictures of the rational expectations hypothesis, whatever they might imply in these situations, each of which appear to be sui generis.
  • Joseph E. Stiglitz, José Antonio Ocampo, Shari Spiegel, Ricardo Ffrench-Davis, and Deepak Nayyar, Stability with Growth. Ch. 6. Open Economy Complications.

• Market valuations can differ substantially and persistently from the rational expectation of the present value of cash flows without leaving statistically discernible traces in the pattern of ex-post returns.
  • Lawrence H. Summers, "Does the Stock Market Rationally Reflect Fundamental Values?" (1986).