Abstract

The development and rationale for GEM can be best explained through a brief description of the role of large macroeconomic policy models. Academic work in macroeconomics tends to focus on specific issues, such as the consumption function or a new theoretical insight. Large macroeconomic policy models, on the other hand, are used to quantify the impact of a range of issues within a unified structure, most notably countercyclical macroeconomic policies. A stylized way of thinking about the interaction between academic work and large policy models is provided in Figure 2.1. A new theoretical insight (such as rational expectations) with strong policy implications is developed in academia in response to evolving policy challenges and the limitations of existing models. Once these ideas have been distilled to the point where they are able to fit the data reasonably, they form the basis for large policy models, starting with single-country versions and then extending to a multicountry setting. Subsequently, the academic and policy communities refine these ideas and the paradigm becomes increasingly dominant. At some point, a new insight emerges and the leading edge of academic work switches to this new paradigm. However, large policy models do not follow because the ideas are not yet able to provide the needed quantitative insights, and academic interest in large macroeconomic models wanes. In short, the “production cycle” of policy models tends to lag that of their academic brethren, given the greater need for policy models to fit the stylized facts of the cycle.

Why a New Model?

The development and rationale for GEM can be best explained through a brief description of the role of large macroeconomic policy models. Academic work in macroeconomics tends to focus on specific issues, such as the consumption function or a new theoretical insight. Large macroeconomic policy models, on the other hand, are used to quantify the impact of a range of issues within a unified structure, most notably countercyclical macroeconomic policies. A stylized way of thinking about the interaction between academic work and large policy models is provided in Figure 2.1. A new theoretical insight (such as rational expectations) with strong policy implications is developed in academia in response to evolving policy challenges and the limitations of existing models. Once these ideas have been distilled to the point where they are able to fit the data reasonably, they form the basis for large policy models, starting with single-country versions and then extending to a multicountry setting. Subsequently, the academic and policy communities refine these ideas and the paradigm becomes increasingly dominant. At some point, a new insight emerges and the leading edge of academic work switches to this new paradigm. However, large policy models do not follow because the ideas are not yet able to provide the needed quantitative insights, and academic interest in large macroeconomic models wanes. In short, the “production cycle” of policy models tends to lag that of their academic brethren, given the greater need for policy models to fit the stylized facts of the cycle.

Figure 2.1.
Figure 2.1.

Stylized View of Model Development

One such major overhaul occurred with the adoption of rational expectations (Table 2.1 summarizes successive generations of policy models). In the 1960s and 1970s large policy models using adaptive expectations and a Keynesian aggregate demand framework quantified the impact of macroeconomic policies. However, in the wake of the great inflation of the 1970s, the implication that output could be raised permanently by injecting aggregate demand through monetary and fiscal policy was recognized as a flaw. Rational expectations fixed this and provided a new range of insights, such as the importance of rules in macroeconomic analysis (Taylor, 1993), exchange rate overshooting (Dornbusch, 1976), and the “random-walk” model of consumption (Hall, 1978). Such models were gradually developed to the point where they could be used in policy circles. Indeed, MULTIMOD, created in the late 1980s, was an early example of a large international version of such a model (see Masson, Symansky, and Meredith, 1990, for a description).

Table 2.1.

Stylized View of the Strengths and Weaknesses of Successive Generations of Macroeconomic Models

article image

These rational expectations models, however, were susceptible to the “Lucas critique.” This was that policy analysis using reduced-form equations that fit the data but were loosely tied to theory—such as those used in large macroeconomic policy models—was fraught with danger, as such models could not adequately account for resulting shifts in behavior (Lucas, 1976). The focus in much of academia in the 1980s and early 1990s was on developing rational expectations models incorporating the explicit microeconomic structure advocated by Lucas. Initially this took the form of “real-business-cycle” models in which prices were assumed to be fully flexible (see Kydland and Prescott, 1982, for a closed economy version andMendoza, 1991, written in the Research Department of the IMF for an open economy one). However, the assumption of flexible prices largely obviated the impact of macroeconomic policies on real activity, making these models of little value in analyzing such policies. Consequently, large policy models generally remained in the reduced-form Keynesian framework, although with an increased focus on adding supply-side linkages.

Over time, it became increasingly clear that the short-term dynamics of real-business-cycle models could be improved by introducing some form of nominal inertia. Theoretical developments in the microeconomics of wage and price setting with imperfect competition led to single-country monetary models that combined the explicit microeconomic foundations typical in real business cycle models with price stickiness (Christiano, Eichenbaum, and Evans, 2001). The leap to multicountry models of this type was accomplished in the mid-1990s (Obstfeld and Rogoff, 1995). The new models merged the microeconomic foundations (of the type advocated by Lucas) with sticky prices, combining production, consumption, nominal rigidities, trade, and international financial markets in a coherent theoretical structure. Work on such models has exploded in recent years (Lane, 2001, provides a survey).

This deep paradigm shift is transforming the study of international finance and international macroeconomics, reevaluating the Mundell-Flemming analysis developed at the Fund in the early 1960s, in the same way that the traditional IS/LM/Phillips curve analysis was recast by recent monetary models (see Obstfeld, 2001). Major insights from these models include that macroeconomic policies, such as the exchange rate regime, can have long-term effects on the level of consumption, labor effort, and the capital stock, in contrast to earlier views generally held in the profession. In addition, policies can be analyzed in terms of their impact on economic welfare of consumers—which includes, for example, the disutility of working harder and having less leisure—rather than their effect on less accurate proxies for welfare such as output and inflation. This has reignited interest in the impact of alternative exchange rate regimes, the benefits from international macroeconomic cooperation, and the role of asset markets in the international business cycle.

Structure of GEM

GEM is a large-scale version of such a micro-founded open economy model. It integrates and builds upon the results in the existing literature—mostly devoted to exploring small and relatively tractable apparatuses—to create a unifying framework for the analysis of international interdependencies. GEM has a modular structure, allowing the model to treat issues in a flexible manner. In addition to GEM, such models have been developed by the Federal Reserve Board (Erceg, Guerrieri, and Gust, 2003), and are areas of active research in several other institutions (such as the central banks of Canada, Finland, Italy, Norway, Spain, and the United Kingdom), and are being considered in some emerging market countries, such as by the central banks in Brazil, Chile, and the Czech Republic. The European Central Bank (ECB) has developed a single-country model (Smets and Wouters, 2002) and is planning a multicountry extension.

The model comprises firms that produce goods, households that consume and provide labor and capital to firms, and a government that taxes and spends (Laxton and Pesenti, 2003). The microeconomic structure of GEM uses standard functional forms that allow firms and consumers to be aggregated as if they were a single entity. On the production side, for example, many small firms produce differentiated goods made using identical constant elasticity of substitution (CES) production functions using labor, capital, and (in some cases) intermediate goods such as components or commodities. Because the goods are differentiated, firms have market power and restrict output to create excess profits. Capital and intermediate goods can be produced and traded while the labor force in each country is fixed, with workers choosing how much to work versus enjoying leisure. Workers also have market power and hence restrict their labor to raise their real wage. The workers own the firms in their country, and hence receive their revenues (net of investment) in the form of wages and profits. This income is spent on home and foreign goods based on a CES utility function. Given the focus on trade and macroeconomic interdependencies, the fiscal and financial sides of the model are currently relatively simple. The government spends on government consumption funded through lump-sum taxes less transfers, domestic financial sectors are not modeled explicitly, while countries pay (receive) a small premium for international borrowing (lending). These sectors are all areas of active development (see Section IV).

To generate realistic dynamics, the model includes judicious use of adjustment costs on real and nominal variables, thereby elongating the responses to shocks and ensuring that consumption and production do not immediately jump to a new long-term equilibrium. On the real side, such costs prolong the adjustment of the capital stock and the level of imports, while “habit persistence” plays a similar role in elongating the responses of consumption and hours worked. Sticky prices are also modeled using adjustment costs, with the prices of domestic goods and imports, as well as wages, displaying inertia. These costs are modeled parsimoniously with only one or two parameters determining the speed of response, and are fully integrated into the theoretical structure.

An innovative feature of GEM compared to most policy models is that it has a flexible structure, so that one can include or exclude features such as non-traded goods, a distribution sector, or trade in commodities or other intermediate goods. In addition, the model can be created with any number of countries, although work to date has involved either two or three countries. Figure 2.2 illustrates the simplest possible version of the two-country model, in which labor and capital are combined to produce a single type of tradable good that can be used for consumption or investment. Given the preferences of consumers, firms, and governments, these goods are then distributed across countries.

Figure 2.2.

Figure 2.3 shows the same two-country model with three major additional features incorporated. The first is that production is split into two stages. In the first stage, labor, capital, and (possibly) land are used to create intermediate goods that can be traded, such as oil or components for manufacturing. These intermediate goods are then combined with additional labor and capital at home and abroad to produce final goods. The addition of intermediate goods allows the model to examine issues that are particularly important for developing countries. These include the policy challenges faced by economies that supply either low value-added components (such as textiles) to industrial countries, assemble higher-technology components from such countries into final products (for example, assembling computers), or are commodity producers and exporters.

Figure 2.3.
Figure 2.3.

More Complicated GEM Structure

Another feature shown in Figure 2.3 is that final goods are split into those that can be traded and those that cannot. Differentiating between traded and nontraded goods is central to a number of issues in international macroeconomics. Most notably, rapid productivity increases in traded goods relative to nontraded goods help explain why real exchange rates tend to appreciate in countries that are growing rapidly—generally referred to as the Balassa-Samuelson effect (Balassa, 1964; and Samuelson, 1964). Including nontraded goods is also useful for many industrial country issues, such as the degree to which actual (and anticipated) productivity increases in information technology goods help explain the strong appreciation of the U.S. dollar over the late 1990s (see Hunt and Rebucci, 2003, for an analysis of this issue using GEM).

Estimating Parameter Values

The deep parameters in micro-founded models like GEM—such as the degrees to which changes in real wages affect the desire to work, consumption responds to changes in real interest rates, firms can substitute labor and capital in response to changing conditions, and home and foreign goods are substitutes—have generally been calibrated using estimates from microeconomic studies combined with an assessment of the way the model fits the overall properties of the data. Unfortunately, parameterization is time consuming, as it involves experimenting across a wide range of potential values. Ideally, it would be better to simplify this process by estimating the parameters from macroeconomic data.

The major constraint to estimating deep parameters is that their impact on the model’s short-term dynamics is often subtle. As the information in the data mainly pertains to these dynamic responses, it is difficult to obtain accurate estimates of crucial elasticities using classical estimation techniques. The standard errors on these coefficients are generally large, and the estimated values often deviate a long way from values regarded as “reasonable” based on theory and microeconomic estimates. Hence the preference for the more time-consuming approach of calibration based on microeconomic evidence.

Providing parameters that fit the facts and are relatively simple to estimate for different countries is particularly important for a policy model such as GEM, as there is a premium on making the model both fit the facts and look at a range of issues across many different countries. Accordingly, the GEM team has been working on using a Bayesian approach to estimate the model’s parameters. Bayesian estimation is the main alternative to the classical approach. The key difference between the two is that in the Bayesian framework the analyst specifies his or her initial view about the value of each parameter and the certainty with which this view is held before estimating the model. This prior information is combined with the evidence contained in the data to obtain final estimates of the model’s parameters. Initial views about parameters can thus be used to improve the accuracy of estimates of key elasticities on which the underlying data have limited information, and the resulting estimates can be used to create parameterizations for simulations.

In many respects, Bayesian estimation is simply a mechanized version of the approach currently used to calibrate models, with its combination of using prior information obtained from outside sources, such as microeconomic studies, with judgments about how well the model fits the data. The great advantage is that, once the Bayesian routine has been established, it can be used to rapidly provide plausible parameterizations across countries. For example, priors can be used to ensure that deep underlying elasticities across a range of countries could be very similar (on the basis that human nature is relatively invariant across countries), while parameters associated with the speed of adjustment, which are more dependent on the particular institutional arrangements in a country, can be derived mainly from the data. Indeed, Bayesian techniques have already been used to generate plausible, albeit preliminary, parameter estimates for a simple closed economy model of the United States, but have not yet been extended to the full GEM.

Finally, a distribution sector is included. There is strong evidence from microeconomic studies that the same goods are sold at different prices across countries. One way of incorporating this observation is to include a distribution sector in the model (Corsetti and Dedola, 2002). All domestic and foreign goods need to go through this sector before they can be bought. As the distribution sector is assumed to consist of nontraded goods, this means that the final prices of all goods are an amalgam of the cost of producing these goods and domestic distribution costs, so prices of imported tradable goods do not fully reflect changes in the real exchange rate even in the long run.

GEM’s flexible modular structure provides a number of advantages. Given the size and complexity of the model, it is often useful to ignore factors that are not of central interest to the issue at hand. For example, while one would wish to include commodities when analyzing a major commodity producer, or the impact of oil price shocks, it is essentially a distraction when looking at countries specialized in manufacturing. Similarly, distribution costs matter when pass-through of exchange rates into prices is important, but is an unnecessary complication for many other issues. In addition, the transmission mechanisms become simpler and more transparent in smaller versions of the model, allowing the consequences of the theoretical structure to be more easily ascertained, making the model less of a “black box.” Simpler models are also essential for some forms of simulation and estimation that are particularly computer intensive.

Once the structure of the model has been determined, the parameters are selected. For most deep parameters that define long-term responses of firms and consumers, such as the responsiveness of hours worked to changes in real wages or the substitutability of different types of goods, estimates from microeconomic studies are used to determine plausible values (Box 2.1 discusses issues associated with estimating parameters). Next, more detailed coefficients are selected to mimic key characteristics of the economic environment, such as the relative size of the countries, their levels of trade, and their capital-output ratios. Finally, the coefficients on costs of adjustment and habit persistence are chosen to generate realistic dynamic responses.

An important feature of any policy model such as GEM is that it fits the dynamics seen in the data. To this end, the adjustment cost parameters are calibrated to fit existing evidence from policy models and estimated vector autoregressions (VARs). Figure 2.4 provides a comparison of the responses to a one-year hike in short-term interest rates in the euro area and the United States in GEM to those from policy models used in central banks—the Area-Wide Model (AWM) of the ECB and the FRB-US model of the Board of Governors of the Federal Reserve System, respectively. These models were chosen as they are primarily designed to fit the dynamics in the data. An alternative is to compare the model to the responses to an interest rate hike found using an estimated vector autoregression (Figure 2.5). In addition to having no imposed theoretical structure, vector autoregressions provide statistical confidence intervals, thereby giving a better sense of the plausibility of the responses produced by GEM.1 In both cases, GEM reproduces the typically hump-shaped path of variables, although the GEM responses tend to be somewhat faster, particularly for investment, possibly reflecting the absence of lags coming from the time it takes to complete a project once initiated. In short, even with its strong theoretical underpinnings, GEM’s structure is rich enough to mimic short-term dynamics.

Figure 2.4.
Figure 2.4.

Dynamic Responses to a One Percentage Point Hike in Interest Rates for One Year: GEM Compared with Large Forecasting Models

(Percent deviation from baseline)

Source: IMF staff calculations based on information provided by the U.S. Federal Reserve Board and the European Central Bank.
Figure 2.5.
Figure 2.5.

Dynamic Responses to a Hike in Interest Rates: GEM Compared with a VAR

Sources: Altig and others (2003) for VAR; and IMF staff calculations for GEM.

Strengths and Weaknesses of GEM

One of the great advantages of GEM compared with earlier types of models is that it can provide evaluations of policies in a general equilibrium setting, thus taking account of the full range of effects across equations. As the model is built from explicit microeconomic foundations, a change in one of the deep parameters in the model can have effects across a wide range of relationships. These complex interrelationships help to identify economic linkages more precisely, providing a stronger framework for analysis that can generate new insights as well as encouraging closer links between IMF researchers and the academic community. This is particularly important at the early stage of creation of a paradigm, when these insights have not been fully incorporated into mainstream analysis, and help explain the enthusiasm GEM has created in academia.

As an example of unveiling linkages, consider a policy that increases competition in the labor market (discussed in more detail in Section III). The most obvious effect is that the market power of workers diminishes, increasing output and, as more goods need to be sold to the rest of the world, depreciating the real exchange rate. The effects on domestic output depend crucially on the response of hours worked to a change in real wages, while the international effects depend on the degree to which home and foreign goods are substitutes. As these parameters are explicitly identified in GEM, the consequences of different assumptions about them can be easily qualified, while in MULTIMOD these elasticities were combined with others in reduced-form relationships. In addition, an important new insight coming from GEM is that more labor market competition reduces nominal inertia. This is because it increases the costs firms incur when wages deviate from their flexible price level. Such an effect could not have been captured in models such as MULTI-MOD with a fixed Phillips curve relationship.

Another example is the impact of industrial country exchange rate volatility on emerging market countries (see IMF, 2003a). The flexibility of GEM means that the effects of structural differences in the emerging market on the impact of such volatility can be explored, including alternative exchange regimes, levels of openness, bilateral trade patterns, levels of debt, and exchange rate pass-through. Insights from this exercise include the importance of the exchange rate regime and degree of domestic exchange rate pass-through on the associated output volatility, effects that depend crucially on the integration of supply, demand, trade, and international asset markets in GEM.

A second advantage of GEM is that the costs and benefits of a policy can be evaluated in a more sophisticated manner. As the model is derived from explicit maximization of profit and utility, one can evaluate policies in terms of their effects on consumer welfare. The advantage of welfare is that it measures the gain to consumers, the ultimate objective of economic activity, and provides a measure of the “best” policy as well as a way of comparing the effects of different policy options. By contrast, in MULTIMOD and similar models benefits were analyzed using more ad hoc measures. For example, monetary policy was generally evaluated in terms of the variability of real output and inflation.

The large size of GEM is an advantage in this respect. Broadly speaking, policies are most useful when they can help reduce economic distortions. As GEM includes a relatively large number of such distortions—such as monopolistic competition, sticky prices, and sluggish adjustment of trade volumes—the potential role for policies to improve welfare is commensurately strengthened. That said, to date it has only been possible to evaluate welfare benefits in small models with simple structures often comprising only one country. This is because of the large computing power needed to solve for a dynamic path with the nonlinear functions needed to calculate welfare. Given the rapid progress in both computing power and solution techniques, however, it is reasonable to expect that full welfare calculations of dynamic simulations will be possible with GEM in the near future.

Even without full welfare analysis, GEM can still provide new insights. As discussed further in Section III, the appropriate monetary rule depends importantly on how potential output is evaluated. A model such as GEM can provide a more accurate calculation of the output gap by incorporating the impact of shocks on the path of aggregate supply. Policy rules using this output gap can then be compared with those using a more traditional approach in which potential output is assumed to change slowly over time, providing insights into the degree to which monetary policymakers should focus on identifying the supply-side implications of disturbances in assessing monetary conditions.

It is important to recognize that moving to a model with a tight theoretical structure also imposes limitations, at least in the short term. Accordingly, MULTIMOD remains a useful tool of analysis, although future development will cease and its use will presumably diminish over time. There are the usual growing pains associated with any new project, such as the need to gain more experience with versions of GEM comprising three or more countries. In addition, the need to create a large interlinked structure constrains theoretical specifications and hence model properties. For example, the use of a representative consumer means that the model is not currently suitable for analysis of income distribution. The need for theoretical consistency can also complicate the addition of new features. For example, the current version of the model does not generate realistic short-term tax multipliers, although this is an area of active development. As discussed further in Section IV, one of the main theoretical approaches to creating such multipliers is to assume that consumers have finite lives. Another implication of finite lives is that consumers’ behavior depends on age and hence cannot be summarized by a single “representative” individual, which creates significant theoretical complications elsewhere.

Finally, calibration of GEM is currently time consuming. This is partly because the concepts in the model often do not dovetail with existing data. For example, it is not easy to split output into traded and nontraded goods or to determine the role of commodities and semifinished components in production. In addition, changes in a coefficient generally affect several equations, with a complex effect on model properties. To date, calibrations have only been completed for three economies—the United States, the euro area, and the Czech Republic. Data sets that will help with calibration have been obtained for over 20 countries and regions, including a wide range of emerging market countries as well as advanced countries, while experience with earlier calibrations are helping to make the process less time consuming.

Cited By

A New International Macroeconomic Model
  • Altig, D., L. Christiano, M. Eichenbaum, and J. Linde, 2003, “Technology Shocks and Aggregate Fluctuations” (unpublished).

  • Balassa, Bela 1964, “The Purchasing-Power Parity Doctrine: A Reappraisal,Journal of Political Economy, Vol. 72, pp. 58496.

  • Bayoumi, Tamim, Douglas Laxton, and Paolo Pesenti, 2004, “Benefits and Spillovers of Greater Competition in Europe: A Macroeconomic Assessment,ECB Working Paper No. 341 (Frankfurt: European Central Bank).

    • Search Google Scholar
    • Export Citation
  • Bernanke, Ben, Mark Gertler, and Simon Gilchrist, 1999, “The Financial Accelerator in a Quantitative Business Cycle,in Handbook of Macroeconomics, ed. by John B. Taylor and Michael Woodford (Amsterdam, New York, and Oxford: Elsevier Science, North-Holland).

    • Search Google Scholar
    • Export Citation
  • Christiano, Lawrence J., Martin Eichenbaum, and Charles Evans, 2001, “Nominal Rigidities and the Dynamic Effects of a Shock to Monetary Policy,NBER Working Paper No. 8403 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation
  • Corsetti, Giancarlo, and Luca Dedola, 2002, “Macroeconomics of International Price Discrimination,Temi de Discussione, No. 461 (Rome: Bank of Italy).

    • Search Google Scholar
    • Export Citation
  • Dornbusch, Rudiger, 1976, “Expectations and Exchange Rate Dynamics,Journal of Political Economy, Vol. 84, No. 6, pp. 116176.

  • Erceg, C., L. Guerrieri, and C. Gust, 2003, “SIGMA: A New Open Economy Model for Policy Analysis,draft paper prepared for the Annual Central Bank Modelers’ Workshop in Amsterdam.

    • Search Google Scholar
    • Export Citation
  • Fleming, J. Marcus, 1962, “Domestic Financial Policies Under Fixed and Under Floating Exchange Rates,Staff Papers, International Monetary Fund, Vol. 9 (November), pp. 36979.

    • Search Google Scholar
    • Export Citation
  • Galí, J., M. Gertler, and J. D. López-Salido, 2002, “Markups, Gaps, and the Welfare Costs of Business Fluctuations,NBER Working Paper No. 8850 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation
  • Hall, Robert E. 1978, “Stochastic Implications of the Life Cycle-Permanent Income Hypothesis: Theory and Evidence,Journal of Political Economy, Vol. 86, No. 6, pp. 97187.

    • Search Google Scholar
    • Export Citation
  • Hart, Oliver, and John Moore, 1994, “A Theory of Debt Based on the Inalienability of Human Capital,Quarterly Journal of Economics, Vol. 109, No. 4, pp. 84179.

    • Search Google Scholar
    • Export Citation
  • Hunt, Benjamin L., and Alessandro Rebucci, 2003, “The U.S. Dollar and the Trade Deficit: What Accounts for the Late 1990s?IMF Working Paper No. 03/194 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2000, “The Impact of Higher Oil Prices on the Global Economy;available via the Internet at http://www.imf.org/external/pubs/ft/oil/2000/oilrep.pdf.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2001, “Business Cycle Linkages Among Major Advanced Economies,World Economic Outlook, October 2001, World Economic and Financial Surveys (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2003a, “How Concerned Should Developing Countries Be About G-3 Exchange Rate Volatility?World Economic Outlook, September 2003, World Economic and Financial Surveys (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2003b, “United States: Selected Issues,IMF Country Report No. 03/245 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Kollman, Robert, 2002, “Monetary Policy Rules in the Open Economy: Effects on Welfare and Business Cycles,CEPR Discussion Paper No. 3279 (London: Centre for Economic Policy Research).

    • Search Google Scholar
    • Export Citation
  • Kydland, Finn E., and Edward C. Prescott, 1982, “Time to Build and Aggregate Fluctuations,Econometrica, Vol. 50 (November), pp. 134570.

    • Search Google Scholar
    • Export Citation
  • Lane, Philip R., 2001, “The New Open Economy Macroeconomics: A Survey,Journal of International Economics, Vol. 54 (August), pp. 23566.

    • Search Google Scholar
    • Export Citation
  • Laxton, Douglas, and Paolo Pesenti, 2003, “Monetary Rules for Small, Open, Emerging Economies,Journal of Monetary Economics, Vol. 50, No. 5, pp. 110946.

    • Search Google Scholar
    • Export Citation
  • Lucas, Robert E., Jr., 1976, “Econometric Policy Evaluation: A Critique,Journal of Monetary Economics, Vol. 1, No. 2, pp. 1946.

    • Search Google Scholar
    • Export Citation
  • Masson, Paul R., Steven A. Symansky, and Guy Meredith, 1990, MULTIMOD Mark II: A Revised and Extended Model, IMF Occasional Paper No. 71 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Mendoza, Enrique G., 1991, “Real Business Cycles in a Small Open Economy,American Economic Review, Vol. 81 (September), pp. 797818.

    • Search Google Scholar
    • Export Citation
  • Mundell, Robert A., 1963, “Capital Mobility and Stabilization Policy Under Fixed and Flexible Exchange Rates,Canadian Journal of Economics and Political Science, Vol. 29, pp. 47585.

    • Search Google Scholar
    • Export Citation
  • Muth, J.F., 1961, “Rational Expectations and the Theory of Price Movements,Econometrica, Vol. 29, pp. 31535.

  • Obstfeld, Maurice, 2001, “International Macroeconomics: Beyond the Mundell-Fleming Model,IMF Staff Papers, Special Issue, Vol. 47, pp. 139.

    • Search Google Scholar
    • Export Citation
  • Obstfeld, Maurice, Kenneth Rogoff, 1995, “Exchange Rate Dynamics Redux,Journal of Political Economy, Vol. 103, pp. 62460.

  • Organization for Economic Cooperation and Development, 1997, The OECD Report on Regulatory Reform, Vol. II: Thematic Studies (Paris: Organization for Economic Cooperation and Development).

    • Search Google Scholar
    • Export Citation
  • Samuelson, Paul A., 1964, “Theoretical Notes on Trade Problems,Review of Economics and Statistics, Vol. 46, pp. 14554.

  • Schadler, Susan M., Paulo F. Drummond, Louis Kuijs, Zuzana Murgasova, Rachel N. van Elkan, 2004, Adopting the Euro in Central Europe: Challenges of the Next Step in European Integration, IMF Ocasional Paper No. 234 (Washington: International Monetary Fund; forthcoming).

    • Search Google Scholar
    • Export Citation
  • Smets, Frank and Raf Wouters, 2002, “An Estimated Stochastic Dynamic General Equilibrium Model of the Euro Area,ECB Working Paper No. 171 (Frankfurt: European Central Bank).

    • Search Google Scholar
    • Export Citation
  • Taylor, John 1993, “Discretion Versus Policy Rules in Practice,Carnegie-Rochester Conference Series on Public Policy, Vol. 39 (December), pp. 195214.

    • Search Google Scholar
    • Export Citation