Article
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

Earlier studies of the seigniorage inflation model have found that the high-inflation steady state is not stable under learning. We reconsider this issue and analyze the full set of solutions for the linearized model. Our main focus is on stationary hyperinflationary paths near the high-inflation steady state. These paths are shown to be stable under least-squares learning if agents can utilize contemporaneous data. In an economy with a mixture of agents, some of whom only have access to lagged data, stable hyperinflationary paths emerge only if the proportion of agents with access to contemporaneous data is sufficiently high.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... The issue of diverging expectations was taken up in the next generation of hyperinflation models, which introduced bounded rationality in order to understand hyperinflationary dynamics; papers such as Barucci (2001), Marcet and Nicolini (2003) , and Adam et al. (2006) relaxed assumptions of rationality in order to explore learning behavior in hyperinflationary environments but still found complex behavior. Adam et al. (2006Adam et al. ( :2743, in particular, touches implicitly upon the chaotic qualities of hyperinflation, noting that if there is limited information globally about other agents in the system, forecasting and expectation-formation becomes highly stochastic. ...
... The issue of diverging expectations was taken up in the next generation of hyperinflation models, which introduced bounded rationality in order to understand hyperinflationary dynamics; papers such as Barucci (2001), Marcet and Nicolini (2003) , and Adam et al. (2006) relaxed assumptions of rationality in order to explore learning behavior in hyperinflationary environments but still found complex behavior. Adam et al. (2006Adam et al. ( :2743, in particular, touches implicitly upon the chaotic qualities of hyperinflation, noting that if there is limited information globally about other agents in the system, forecasting and expectation-formation becomes highly stochastic. This result comports both with real-world empirics of hyperinflationary paths and with economic intuition regarding the (lack of) universal availability of knowledge to economic agents at even the local level ( Hayek, 1945 ). ...
Article
Full-text available
Monetary policies follow a complex and chaotic process, one that is enhanced when monetary growth is taken to the extreme, as in hyperinflation. While we have a basic understanding of the complex dynamics of hyperinflation, a less-explored phenomenon accompanying hyperinflationary episodes is the effect that they have on institutional development. Building on recent advances in economic history examining monetary policy and institutions, this analysis uses more explicit recourse to complexity theory and in particular complex adaptive systems to examine two separate hyperinflationary episodes in Hungary, 1923 and 1946. Delving into the institutional and cultural roots of the inflationary policies of the Hungarian government and detailing the institutional effects which succeeded hyperinflation, this paper shows that the twin experiences of Hungary were both chaotic and created a chaotic and complex aftermath, the results of which are still being felt. Indeed, Hungary's 1923 and especially its 1946 hyperinflation shows that some economic chaos cannot be restrained by political institutions and in fact the institutions themselves were changed. When staring into the abyss, Hungary also found that the abyss stared back.
... This work has been interpreted as supporting the notion that low in ‡ation, monetary equilibria are attractors under adaptive learning processes in overlapping generations models which are known to admit multiple equilibria. More recently, Lettau and Van Zandt (2001) and Adam et al. (2006) have shown in the seigniorage in ‡ation overlapping generations monetary model that the high in ‡ation steady state (Lettau and Van Zandt (2001)) or stationary paths near that steady state (Adam et al. (2006)) may be stable under adaptive learning dynamics under certain restrictive timing assumptions, e.g., if agents have contemporary observations of endogenous variables in the information sets they use to form future expectations. These …ndings cast some doubt on Lucas's suggestion that adaptive learning dynamics might provide a means of selecting between the low and high in ‡ation stationary equilibria of the model as it appears that under certain conditions both equilibria might be learnable. ...
... This work has been interpreted as supporting the notion that low in ‡ation, monetary equilibria are attractors under adaptive learning processes in overlapping generations models which are known to admit multiple equilibria. More recently, Lettau and Van Zandt (2001) and Adam et al. (2006) have shown in the seigniorage in ‡ation overlapping generations monetary model that the high in ‡ation steady state (Lettau and Van Zandt (2001)) or stationary paths near that steady state (Adam et al. (2006)) may be stable under adaptive learning dynamics under certain restrictive timing assumptions, e.g., if agents have contemporary observations of endogenous variables in the information sets they use to form future expectations. These …ndings cast some doubt on Lucas's suggestion that adaptive learning dynamics might provide a means of selecting between the low and high in ‡ation stationary equilibria of the model as it appears that under certain conditions both equilibria might be learnable. ...
Article
Adaptive learning is important in dynamic models since it is a process that shows the improvement in the understanding of the agents of the model. Whenever there is a dynamic environment, there is a room for improvement through learning. In this thesis I analyze the adaptive learning of the agents in different setups. In my first paper I show that adaptive learning does not eliminate the multiplicity of stationary equilibria in the Diamond overlapping generations model with money and productive capital; both dynamically efficient and inefficient equilibria are found to be stable under adaptive learning. In my second paper I show that the two agents of a natural-rate model, with different beliefs, learn the economy which leads to convergence or endogenous fluctuations of the inflation rate under different conditions. And in my last paper I show that a central bank with an extraneous instrument, "cheap talk" announcements, can influence the private sector to achieve better outcomes than could be obtained by manipulating the nominal interest rate alone with full knowledge of private sector expectation formation and in anything less than full knowledge, the private sector learns to discount announcements.
... 2 0 For an overview of these methods, see Evans and Honkapohja (2001) and Hommes (2013). 2 1 In an earlier version of the paper, we allowed a fraction 2 [0; 1) of agents in the CE model to make use of contemporaneous information about the exchange rate, similar to the information setup in Adam et al. (2006). The quantitative results were broadly similar to those presented here. ...
... For an overview of these methods, seeEvans and Honkapohja (2001) andHommes (2013).21 In an earlier version of the paper, we allowed a fraction λ ∈ [0, 1) of agents in the CE model to make use of contemporaneous information about the exchange rate, similar to the information setup inAdam et al. (2006). The quantitative results were broadly similar to those presented here. ...
... The obvious problem when tackling these issues empirically is 1 Recent contributions include Williams (2005) and Adam et al. (2006). to measure expected and unexpected ination. Nevertheless, there is a rich literature based on survey data. ...
Article
Full-text available
The present paper compares expected inflation to (econometric) inflation forecasts based on a number of forecasting techniques from the literature using a panel of ten industrialized countries during the period of 1988 to 2007. To capture expected inflation we develop a recursive filtering algorithm which extracts unexpected inflation from real interest rate data, even in the presence of diverse risks and a potential Mundell-Tobin-effect. The extracted unexpected inflation is compared to the forecasting errors of ten econometric forecasts. Beside the standard AR(p) and ARMA(1,1) models, which are known to perform best on average, we also employ several Phillips curve based approaches, VAR, dynamic factor models and two simple model avering approaches.
Article
Full-text available
Monetary policies follow a complex and chaotic process, one that is enhanced when monetary growth is taken to the extreme, as in hyperinflation. While we have a basic understanding of the complex dynamics of hyperinflation, a less-explored phenomenon accompanying hyperinflationary episodes is the effect that they have on institutional development. Building on recent advances in economic history examining monetary policy and institutions, this analysis uses more explicit recourse to complexity theory and in particular complex adaptive systems to examine two separate hyperinflationary episodes in Hungary, 1923 and 1946. Delving into the institutional and cultural roots of the inflationary policies of the Hungarian government and detailing the institutional effects which succeeded hyperinflation, this paper shows that the twin experiences of Hungary were both chaotic and created a chaotic and complex aftermath, the results of which are still being felt. Indeed, Hungary’s 1923 and especially its 1946 hyperinflation shows that some economic chaos cannot be restrained by political institutions and in fact the institutions themselves were changed. When staring into the abyss, Hungary also found that the abyss stared back.
Article
This paper reevaluates the basic prescriptions of monetary policy design in the new Keynesian paradigm through the lens of imperfect knowledge. We show that while the basic logic of monetary policy design under rational expectations continues to obtain, perfect knowledge and learning can limit the set of policies available to central banks, rendering expectations management in general more difficult. Nonetheless, the desirability of some form of price-level targeting, inducing inertia in interest-rate policy, paramount under rational expectations, is robust to the assumption of imperfect knowledge.
Article
This study explores the macroeconomic implications of adaptive expectations in a standard real business cycle model. When rational expectations are replaced by adaptive expectations, we show that the self-confirming equilibrium is the same as the steady-state rational expectations equilibrium for all admissible parameters but that dynamics around the steady state are substantially different between the two equilibria. The differences are driven mainly by the dampened wealth effect and the strengthened intertemporal substitution effect, not by the escapes emphasized by Williams (2003). As a result, adaptive expectations can be an important source of frictions that amplify and propagate technology shocks and seem promising for generating plausible labor market dynamics.
Chapter
The paper aims at assessing the forecast risk and the maximum admissible forecast horizon for the non-systematic component of inflation modeled autoregressively, where a distortion is caused by a simple nonlinear (bilinear) process. The concept of the guaranteed upper risk of forecasting and the δ-admissible distortion level is defined. In order to make this concept operational we propose a method of evaluation of the p-maximum admissible forecast risk, on the basis of the maximum likelihood estimates of the bilinear coefficient. It has been found that for the majority of developed countries (in terms of average GDP per capita) the maximum admissible forecast horizon is between 5 and 12 months, while for the poorer countries it is either shorter than 5 or longer than 12. There is also a negative correlation of the maximum admissible forecast horizon with the average GDP growth.
Article
This paper introduces imperfect knowledge and learning behavior of economic agents into the Kiyotaki and Moore model and studies the interaction of agents' collateral price beliefs, collateral constraint, and aggregate economic activity over the business cycle. It establishes the E-stability condition and the convergence of the real time learning process. In addition, it shows that learning strengthens the role of collateral constraints in aggregate fluctuations.
Article
This paper investigates the learnability of an equilibrium with private information. Agents of each type have their own private information about an exogenous variable and conduct adaptive learning with a heterogeneously misspecified perceived laws of motion (PLM) that includes only this variable. The paper shows that the existence of private information has a nonnegative impact on the learnability of the equilibrium; that is, the condition for learnability is unaffected or relaxed by heterogeneity and/or misspecification in PLMs caused by private information. In a New Keynesian model with private information about fundamental shocks, the learnability of the equilibrium is ensured by the Taylor principle of monetary policy. The paper also confirms that these results hold true not only in the presence of private information, but also in a variety of informational structures.
Article
This paper characterizes optimal fiscal policy when agents learn about future taxation. A benevolent and fully rational government chooses taxes on labor income and state-contingent bonds to finance public spending, considering that private agents form their expectations through a learning algorithm. Facing a trade-off between distortionary taxes and distorted expectations, the Ramsey planner chooses the policy that minimizes the total cost of distortions. The analysis produces two main results. First, the government will use fiscal variables to manipulate expectations, reducing taxes and issuing debt at times of pessimism and doing the opposite at times of optimism. This speeds up learning. Second, the expectation-dependent fiscal plan is also history-dependent, and it prescribes taxes that are not as smooth and more persistent than under rational expectations. These findings are robust to alternative learning algorithms.
Article
This paper investigates three classic questions in monetary theory: How can an intrinsically worthless asset, such as fiat money, maintain value as a medium of exchange? What are the short-run and long-run effects of a change in the money supply? What is the social cost of inflation? I answer these questions using a microfounded model of monetary exchange that replaces the rational expectations assumption with an adaptive learning rule. First, I show that monetary exchange is a robust arrangement in the sense that agents are able to learn the stationary monetary equilibrium while the non-monetary equilibrium is unstable under learning. Second, an unanticipated monetary injection has real effects in the short-run because learning the value of money takes time. In the long run, agents successfully learn the value of money, hence money is neutral. Third, under a constant money growth policy, an increase in the growth rate of money increases output in the short-run producing a short-run Phillips curve. A ten percent increase in the money growth rate has a social cost of 0.41 percent of output per year. Alternatively, a ten percent decrease in the money growth rate has a social benefit of 0.37 percent of output per year.
Article
An adaptive step-size algorithm [Kushner and Yin, Stochastic Approximation and Recursive Algorithms and Applications, 2nd ed., New York: Springer-Verlag (2003)] is used to model time-varying learning, and its performance is illustrated in the environment of Marcet and Nicolini [American Economic Review 93 (2003), 1476–1498]. The resulting model gives qualitatively similar results to those of Marcet and Nicolini, and performs quantitatively somewhat better, based on the criterion of mean squared error. The model generates increasing gain during hyperinflations and decreasing gain after hyperinflations end, which matches findings in the data. An agent using this model behaves cautiously when faced with sudden changes in policy, and is able to recognize a regime change after acquiring sufficient information.
Article
Full-text available
We consider the stability under adaptive learning dynamics of steady state equilibria in Diamond`s (1965) overlapping generations growth model with capital and money. Interior steady state equilibria of this model can be either dynamically inefficient or dynamically efficient. We show that a necessary condition for an equilibrium of this model to be stable under adaptive learning is that the equilibrium is dynamically efficient. In other words, adaptive learning can be used as a selection criterion to exclude dynamically inefficient equilibria. We also provide conditions under which a dynamically efficient equilibrium of this model involving the use of both capital and money will be stable under adaptive learning dynamics.
Article
The present paper compares expected inflation to (econometric) inflation forecasts based on a number of forecasting techniques from the literature using a panel of ten industrialized countries during the period from 1988 to 2007. To capture expected inflation, we develop a recursive filtering algorithm that extracts unexpected inflation from real interest rate data, even in the presence of diverse risks and a potential Mundell–Tobin-effect.The extracted unexpected inflation is compared to the forecasting errors of ten econometric forecasts. In addition to the standard AR(p) and ARMA(1,1) models, which are known to have the best performance on average, we also employ several Phillips curve-based approaches, VAR, dynamic factor models and two simple model averaging approaches.Finally, we show that the quality of the expectations clearly matches the quality of the forecasts derived with the techniques that are currently proposed for this purpose in the economic literature.Highlights► We develop a new panel filter to identify inflation expectations. ► We compare expectations to a set of econometric models of inflation. ► Expectations perform well in terms of precision compared to all benchmark models. ► Expectations include information that is not included in most forecasting models. ► However, expectations do not exploit all available information.
Article
We propose a rational expectations framework for understanding speculative hyperinflations that end in response to 'orthodox' stabilization programs. Motivated by a strong degree of hysteresis in the stock of real balances after the end of hyperinflations, we provide a cash-and-credit model in which the money demand exhibits persistence because individuals can establish long-lasting credit relationships. We use the model to show that if hysteresis in real balances is possible then a fiscal-monetary reform that successfully stops a speculative hyperinflation may fail to prevent it. We argue that speculative hyperinflationary equilibria are consistent with some key stylized facts observed in extreme hyperinflations.
Article
We derive the optimal fiscal policy for a government which is committed to honour its debts but faces investors which fear that a sovereign default might instead happen. We assume that investors are able to learn from new evidence, as in Marcet and Sargent (1989), so that they can correct over time their overly pessimistic view about government?s creditworthiness. We show that in this economy, contrary to the prescriptions of standard models, a frontloaded fiscal consolidation after an adverse fiscal shock is optimal.
Article
This paper presents a linear self-referential macroeconomic model with the possibility of multiple equilibria where agents have the choice of using one of two forecasting models (one of minimum state variable form and the other of sunspot form) to form expectations of current and future prices. Endogenous predictor selection is modeled as an evolutionary game where individuals choose among the forecasting models based on relative performance. Some Nash solutions are not relevant as they are not stable under evolutionary or adaptive learning. Finally, it is shown that the sunspot equilibrium is fragile against temporary shocks to information costs.
Article
We infer determinants of Latin American hyperinflations and stabilizations by using the method of maximum likelihood to estimate a hidden Markov model that assigns roles both to fundamentals in the form of government deficits that are financed by money creation and to destabilizing expectations dynamics that can occasionally divorce inflation from fundamentals. Levels and conditional volatilities of monetized deficits drove most hyperinflations and stabilizations, with a notable exception in Peru, where a cosmetic reform of the type emphasized by Marcet and Nicolini occurred. (c) 2009 by The University of Chicago. All rights reserved..
Article
Parallel computation has a long history in econometric computing, but is not at all wide spread. We believe that a major impediment is the labour cost of coding for parallel architectures. Moreover, programs for specific hardware often become obsolete quite quickly. Our approach is to take a popular matrix programming language (Ox), and implement a message-passing interface using MPI. Next, object-oriented programming allows us to hide the specific parallelization code, so that a program does not need to be rewritten when it is ported from the desktop to a distributed network of computers. Our focus is on so-called embarrassingly parallel computations, and we address the issue of parallel random number generation.
Article
The paper emerges from the failure of the traditional models of hyperinflation with rational expectations or perfect foresight. Using the insights from two standard optimizing monetary settings the paper shows that the possibility of perfect foresight monetary hyperinflation paths depends robustly on the essentiality of money. We show that the popular semilogarithmic form of the demand for money is not appropriate to analyse monetary hyperinflation with perfect foresight. We propose a simple test of money essentiality for the appropriate specification of the demand for money equation in empirical studies of hyperinflation.
Article
The assumption of perfectly rational representative agents is now commonly questioned. This paper explores the equilibrium properties of boundedly rational heterogeneous agents. We combine an adaptive learning process in a modified cobweb model within a Stackleberg framework. We assume that there is an asymmetric information diffusion process from leading to following firms. In contrast to a simple cobweb model which has a unique REE, our model may produce multiple restricted perceptions equilibria (RPE). However, a unique and learnable RPE, under certain conditions, can exist in our model. In addition, the following firms' forecasts can confound the leading firms' forecasts -- when the following firms misinterpret information coming from the leading firms. We refer this situation to the boomerang effect. We also find that the leading firms' mean squared forecast error can be even larger than that of following firms if the proportion of following firms is sufficiently large in the market
Article
This study explores the macroeconomic implications of adaptive expectations in a standard growth model. We show that the self-confirming equilibrium under adaptive expectations is the same as the steady state rational expectations equilibrium for all admissible parameter values, but that dynamics around the steady state are substantially different between the two equilibria. The differences are driven mainly by the dampened wealth effect and the strengthened intertemporal substitution effect, not by escapes emphasised by Williams (2003) . Consequently, adaptive expectations can be an important source of frictions that amplify and propagate technology shocks and seem promising for generating plausible labour market dynamics. Copyright © The Author(s). Journal compilation © Royal Economic Society 2009.
Article
This paper shows that the concept of Expectational stability (E-stability) in a multivariate framework is inherently model-dependent. Whereas a Rational Expectations equilibrium (REE) is subject to model-specific parameter restrictions from the economic model at hand, a perceived law of motion (PLM) is postulated without such restrictions because economic agents are not likely to know the restrictions a priori. Therefore, an unrestricted PLM is in general overparameterized relative to an REE of interest in multivariate models even when the functional form is the same as the REE. Since E-stability necessarily involves model-specific extents of overparameterization, it is model-dependent in general. An immediate implication is that E-stability in a multivariate framework is not directly comparable across models and, in particular, across different representations of a given model. This implies that one may draw different conclusions on E-stability of an REE to one model under alternative representations of the model and the REE. We discuss a potential direction to develop a model-independent concept of E-stability.
Article
This study explores theoretical and macroeconomic implications of the self-confirming equilibrium in a standard growth model. When rational expectations are replaced by adaptive expectations, we prove that the self-confirming equilibrium is the same as the steady state rational expectations equilibrium, but that dynamics around the steady state are substantially different between the two equilibria. We show that, in contrast to Williams (2003), the differences are driven mainly by the lack of the wealth effect and the strengthening of the intertemporal substitution effect, not by escapes. As a result, adaptive expectations substantially alter the amplification and propagation mechanisms and allow technology shocks to exert much more impact on macroeconomic variables than do rational expectations.
Article
Full-text available
This paper explores the equilibrium properties of boundedly rational heterogeneous agents under adaptive learning. In a modified cobweb model with a Stackelberg framework, there is an asymmetric information diffusion process from leading to following firms. It turns out that the conditions for at least one learnable equilibrium are similar to those under homogeneous expectations. However, the introduction of information diffusion leads to the possibility of multiple equilibria and can expand the parameter space of potential learnable equilibria. In addition, the inability to correctly interpret expectations will cause a on the forecasts and forecast efficiency of the leading firms. The leading firms' mean square forecast error can be larger than that of following firms if the proportion of following firms is sufficiently large.
Article
Full-text available
Expectations play a central role in modern macroeconomic theories. The econometric learning approach models economic agents as forming expectations by estimating and updating forecasting models in real time. The learning approach provides a stability test for rational expectations and a selection criterion in models with multiple equilibria. In addition, learning provides new dynamics if older data is discounted, models are misspecified or agents choose between competing models. This paper describes the E-stability principle and the stochastic approximation tools used to assess equilibria under learning. Applications of learning to a number of areas are reviewed, including the design of monetary and fiscal policy, business cycles, self-fulfilling prophecies, hyperinflation, liquidity traps, and asset prices.
Article
Expectations about the future are central for determination of current macroeconomic outcomes and the formulation of monetary policy. Recent literature has explored ways for supplementing the benchmark of rational expectations with explicit models of expectations formation that rely on econometric learning. Some apparently natural policy rules turn out to imply expectational instability of private agents’ learning. We use the standard New Keynesian model to illustrate this problem and survey the key results for interest-rate rules that deliver both uniqueness and stability of equilibrium under econometric learning. We then consider some practical concerns such as measurement errors in private expectations, observability of variables and learning of structural parameters required for policy. We also discuss some recent applications, including policy design under perpetual learning, estimated models with learning, recurrent hyperinflation, and macroeconomic policy to combat liquidity traps and deflation.
Article
Expectations of the future play a large role in macroeconomics. The rational expectations assumption, which is commonly used in the literature, provides an important benchmark, but may be too strong for some applications. This paper reviews some recent research that has emphasized methods for analyzing models of learning, in which expectations are not initially rational but which may become rational eventually provided certain conditions are met. Many of the applications are in the context of popular models of monetary policy. The goal of the paper is to provide a largely nontechnical survey of some, but not all, of this work and to point out connections to some related research.
Article
Full-text available
Expectations about the future are central for determination of current macroeconomic outcomes and the formulation of monetary policy. Recent literature has explored ways for supplementing the benchmark of rational expectations with explicit models of expectations formation that rely on econometric learning. Some apparently natural policy rules turn out to imply expectational instability of private agents' learning. We use the standard New Keynesian model to illustrate this problem and survey the key results about interest-rate rules that deliver both uniqueness and stability of equilibrium under econometric learning. We then consider some practical concerns such as measurement errors in private expectations, observability of variables and learning of structural parameters required for policy. We also discuss some recent applications including policy design under perpetual learning, estimated models with learning, recurrent hyperinflations, and macroeconomic policy to combat liquidity traps and deflation.
Article
This paper introduces a general method to study learnability of equilibria resulting from agents using misspecified forecasting models. One can represent the actual and perceived laws of motion (PLM) as seemingly unrelated regressions and then linearly project the actual law of motion into the same class as the PLM. I present an application using the New Keynesian IS-AS model with inertia under several simple Taylor policy rules. It turns out that the results presented in Bullard and Mitra [2002. Learning about monetary policy rules. Journal of Monetary Economics 49, 1105-1129; 2005. Determinacy, Learnability, and Monetary Policy Inertia. Journal of Money, Credit, and Banking, forthcoming] are robust when agents do not include all the state variables in their forecasting models.
Article
Business cycles in different regions of the United States tend to synchronize. This study investigates the reasons behind this synchronization of business cycles and the consequent formation of a national business cycle. Trade between regions may not be strong enough for one region to "drive" business cycle fluctuations in another region. This study suggests that regional business cycles synchronize due to a nonlinear "mode-locking" process in which weakly coupled oscillating systems (regions) tend to synchronize. There is no definitive test for mode-lock. However, simulations, correlations, Granger causality tests, tests for nonlinearities, vector autoregressions, and spectral analysis reveal modest econometric support for the regional mode-locking hypothesis of business cycle synchronization. Copyright Blackwell Publishers, 2005
Article
Motivated by a strong degree of hysteresis in the stock of monetization observed after the end of hyperinflations, I provide a cash-and-credit model in which the use of money exhibits some persistence because individuals can establish long-lasting credit relationships. This feature helps to account for the main stylized facts of extreme hyperinflations and reconcile some conflicting views on their causes, development and end without departing from rational expectations. Unlike the existing literature, I show that when hysteresis is possible, an orthodox fiscal-monetary reform that successfully stops a speculative hyperinflation may not be sufficient to prevent it.
Article
Full-text available
Many macroeconomic models involving rational expect at ions give rise to an infinity of solution paths, even when the models are linear in all variables. Some writers have suggested that this non-uniqueness constitutes a serious weakness for the rational expectations hypothesis. One purpose of the present paper is to argue that the non-uniqueness in question is not properly attributable to the rationality hypothesis but, instead, is a general feature of dynamic models involving expectations. It is also argued that there typically exists, in a very wide class of linear rational expectations models, a single solution that excludes "bubble" or "bootstrap" effects -- ones that occur only because they are arbitrarily expected to occur. A systematic procedure for obtaining solutions free from such effects is introduced and discussed. In addition, this procedure is used to interpret and reconsider several prominent examples with solution multiplicities, including ones developed by Fischer Black and John B. Taylor.Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.
Article
Full-text available
Inflation and the monetary financing of deficits are analyzed in a model in which the deficit is constrained to be less than a given fraction of a measure of aggregate market activity. Depending on parameter values, the model can have multiple steady states. Under adaptive learning with heterogeneous learning rules, there is convergence to a subset of these steady states. In some cases, a high-inflation constrained steady state will emerge. However, with a sufficiently tight fiscal constraint, the low-inflation steady state is globally stable. We provide experimental evidence in support of our theoretical results.
Article
Full-text available
We study adaptive learning in a monetary overlapping generations model with sticky prices and monopolistic competition for the case where learning agents observe current endogenous variables. Observability of current variables is essential for informational consistency of the learning setup with the model setup but generates multiple temporary equilibria when prices are flexible and prevents a straightforward construction of the learning dynamics. Sticky prices overcome this problem by avoiding simultaneity between prices and price expectations. Adaptive learning then robustly selects the determinate (monetary) steady state independent from the degree of imperfect competition. The indeterminate (non-monetary) steady state and non-stationary equilibria are never stable. Stability in a deterministic version of the model may differ because perfect foresight equilibria can be the limit of restricted perceptions equilibria of the stochastic economy with vanishing noise and thereby inherit different stability properties. This discontinuity at the zero variance of shocks suggests to analyze learning in stochastic models.
Article
Full-text available
We study adaptive learning in a monetary overlapping generations model with sticky prices and monopolistic competition for the case where learning agents observe current endogenous variables. Observability of current variables is essential for informational consistency of the learning setup with the model setup but generates multiple temporary equilibria when prices are flexible and prevents a straightforward construction of the learning dynamics. Sticky prices overcome this problem by avoiding simultaneity between prices and price expectations. Adaptive learning then robustly selects the determinate (monetary) steady state independent from the degree of imperfect competition. The indeterminate (non-monetary) steady state and non-stationary equilibria are never stable. Stability in a deterministic version of the model may differ because perfect foresight equilibria can be the limit of restricted perceptions equilibria of the stochastic economy with vanishing noise and thereby inherit different stability properties. This discontinuity at the zero variance of shocks suggests one should analyse learning in stochastic models.
Chapter
In Sargent and Wallace (1973), we created a rational expectations model of the bivariate inflation-money creation process by solving the ‘inverse optimal predictor problem’ for Cagan’s (1956) adaptive expectations scheme. Our model was constructed by working backwards and finding a bivariate process for inflation and money creation that implied both that Cagan’s portfolio balance equation held and that adaptive expectations were rational.1 That model is ‘successful’ in several ways. First, the model explains the pattern of Granger causality in Cagan’s data, in which inflation Granger causes money creation, but not vice versa. Second, a version of the model predicts the pattern of correlations across countries between Cagan’s estimates of X and a (see Sargent and Wallace, 1973). Third, the model predicts that the residuals in the regression equation fit by Cagan will be random walks, which explains the very high serial correlation that Cagan actually encountered (see Sargent, 1977). Fourth, the model predicts a pattern of results obtained by Rodney Jacobs (1975) when he reversed the direction of regression in Cagan’s equation (see Sargent, 1976).
Article
This paper shows how to use an apparatus of A. Marcet and the author [see J. Econ. Theory 48, No.2, 337-368 (1989; Zbl 0672.90023); J. Polit. Econ. 97, 1306-1322 (1989)] to compute equilibria of a class of models described by R. M. Townsend [ibid. 91, 546-588 (1983)]. An equilibrium is modelled in which agents are extracting signals from observations on endogenous variables. By modelling agents’ perceived laws of motion as vector ARMA processes, the equilibrium of the model can be formulated as the fixed point of an operator that maps perceived laws of motion into statistically optimal estimators of those laws of motion.
Article
This paper studies overlapping generations economies in which agents use genetic algorithms to learn correct decision rules. The results of computer simulations show that a genetic algorithm converges to the unique monetary steady state in case of a constant money supply policy and to the low-inflation stationary equilibrium in case of a constant real deficit financed through seignorage. Features of the genetic algorithm adaptation are compared to the performance of other learning algorithms and to the behavior observed in experiments with human subjects in the same OLG environments.
Article
This paper examines disequilibrium adaptive learning behavior in an overlapping generations model with fiat money. Agents are concerned with forming correct forecasts of future inflation. If they use a disequilibrium, adaptive forecast rule, it is shown that they will eventually learn to believe in a nonstationary, nonunique perfect foresight equilibrium. The nonstationary equilibrium isolated by the adaptive learning process can be used to explain the sluggish adjustment of the price level to monetary disturbances as documented in the time series analysis of Sims (Amer. J. Agr. Econ.71 (1989), 489-494). Journal of Economic Literature Classification Numbers: D83, E31.
Article
Parallel computation has a long history in econometric computing, but is not at all wide spread. We believe that a major impediment is the labour cost of coding for parallel architectures. Moreover, programs for specific hardware often become obsolete quite quickly. Our approach is to take a popular matrix programming language (Ox), and implement a message-passing interface using MPI. Next, object-oriented programming allows us to hide the specific parallelization code, so that a program does not need to be rewritten when it is ported from the desktop to a distributed network of computers. Our focus is on so-called embarrassingly parallel computations, and we address the issue of parallel random number generation.
Article
The concept of economic reform is described as a planned shift from one Pareto inefficient, but quasi-stable, Nash equilibrium (or "trap") to a new Pareto superior equilibrium, which will also be stable. The concept is applied to recent "shock" stabilization programs, with special reference to Israel, where the economy was credibly shifted from a three-digit inflationary process, with considerable inertia, to relative price stability with higher real growth, at only moderate adjustment costs, by means of a "heterodox" plan. The idea is rationalized with a simple dual equilibrium inflation model, for which some econometric estimates are also given. Copyright 1989 by The Econometric Society.
Article
This paper studies the selection of information collecting agents by policy makers in the light of two agency problems. First, it is often hard to ascertain how much effort agents have put in acquiring information. Second, when agents have an interest in the policy outcome, they may manipulate information. We show that unbiased advisers put highest effort in collecting information. Eliminating manipulation of information, however, requires that the preferences of the policy maker and the adviser be aligned. Therefore, policy makers appoint advisers with preferences that are less extreme than their own. Copyright 2005 Royal Economic Society.
Article
The paper opens with a description of the salient features of the Israeli economy. These consist of a large government sector(the government budget has absorbed more than 80% of GNP in some recent years); high levels of defense spending; a large government budget deficit; a large current account deficit (about 20%of GNP); triple digit inflation; and extensive indexation of both wages and long term financial commitments. A descriptive model of the economy is then presented, which includes the particular asset menu of the Israeli economy, and its properties examined. Finally,the rrodel is used in analyzing aspects of the Israeli inflationary experience.The currency liberalization of 1977, which increased the access of Israelis to foreign assets, shares respensibility for the high rate of inflation. The possibilities of ending the inflation are discussed.
Article
Dynamic models in which agents' behaviour depends on expectations of future prices or other endogenous variables can have steady states that are stationary equilibria for a wide variety of expectations rules, including rational expectations. When there are multiple steady states, stability is a criterion for selecting predictions of long-run outcomes among them. The purpose of this Paper is to study how sensitive stability is to certain details of the expectations rules, in a simple OLG model with constant government debt that is financed through seigniorage. We compare simple recursive learning rules, learning rules with vanishing gain, and OLS learning, and also relate these to expectational stability. One finding is that two adaptive expectation rules that differ only in whether they use current information can have opposite stability properties.
Article
This paper uses a model of boundedly rational learning to account for the observations of recurrent hyperinflations in the last decade. We study a standard monetary model, where the full rational expectations assumption is replaced by a formal definition of quasi-rational learning. The model under learning is able to match remarkably well some crucial stylized facts, observed during the recurrent hyperinflations experienced by several countries in the 1980s. We argue that, despite being a small departure from rational expectations, quasi-rational learning does not preclude falsifiability of the model and it does not violate reasonable rationality requirements.
Article
The authors design and study an OLG experimental economy where the government finances a fixed real deficit through seigniorage. The economy has continua of nonstationary rational expectations equilibria and two stationary rational expectations equilibria. The authors do not observe nonstationary rational expectations paths. Observed paths tend to converge close to, or somewhat below, the low inflation stationary state. The adaptive learning hypothesis is consistent with the data in selecting the low inflation stationary state rational expectations equilibrium as a long-run stationary equilibrium. Nevertheless, simple adaptive learning models do not capture the market uncertainty or the biases observed in the data. Copyright 1993 by The Econometric Society.
The monetary dynamics of hyper-inflation. In: Friedman (Ed.), Studies in the Quantity Theory of Money On learning and the nonuniqueness of equilibrium in an overlapping generations model with fiat money
  • Article In
  • P Cagan
ARTICLE IN PRESS Cagan, P., 1956. The monetary dynamics of hyper-inflation. In: Friedman (Ed.), Studies in the Quantity Theory of Money. University of Chicago Press, Chicago. Duffy, J., 1994. On learning and the nonuniqueness of equilibrium in an overlapping generations model with fiat money. Journal of Economic Theory 64, 541–553.
Equilibrium with Signal Extraction from Endogenous Variables Inflation and the Government Budget Constraint
SS, T. J. (1991): " Equilibrium with Signal Extraction from Endogenous Variables, " Journal of Economic Dynamics and Control, 15, 245—273. SS, T. J., N. W (1987): " Inflation and the Government Budget Constraint, " in (Razin and Sadka 1987).
Studies in the Quantity Theory of Money Convergence of Least Squares Learning and the Dynamic of Hyperinflation
  • M Chicago
  • A Mm
  • T J Ss
FF, M. (ed.) (1956): Studies in the Quantity Theory of Money. University of Chicago Press, Chicago. MM, A., T. J. SS (1989): " Convergence of Least Squares Learning and the Dynamic of Hyperinflation, " in (Barnett, Geweke, and Shell 1989), pp. 119—137.