Research program
Monetary and Fiscal Stabilization policies, the exchange rate and the term structure of interest rates
University Co-ordinator
Università degli Studi di BOLOGNA -
SCIENZE ECONOMICHE - BOLOGNA(BO)
Research Unit Leader
Giorgio BASEVI
Description
The Unit of Bologna will articulate the research program over three main areas, (i) monetary and fiscal policy interactions within Dynamic Stochastic Models; (ii) Heterogeneous expectations and monetary and fiscal policy interactions; solution determinacy and E-stability conditions for the Rational Expectations Equilibrium; (iii) corporate and government bonds indexation models. (i) The first topic – developed jointly with the Unit of Milano Bicocca and Bocconi –concerns the analysis of the conditions determining an optimal monetary policy rule conditioned to a variety of fiscal policy rule defined according to FTPL. In this context, we plan to examine monetary policy rules conditionally to two different fiscal policy rules belonging to the FTPL class: the first is the standard fiscal policy reaction function highlighted by Leeper (1991) where fiscal revenue is set to react proportionally to the outstanding stock of real debt. A second rule instead, will be designed in order to determine the fiscal revenue as a function of the burden of interest expenses.The analytical context to be developed is represented by a Dynamic Stochastic General Equilibrium model with specific characteristics designed in order to realize the inertial pattern of inflation and real variables. In particular, we are going to insert nominal and real rigidities under the form of price and wage adjustment costs, together with real rigidities, under the form of quadratic cost of capital adjustments. For a better characterization of the inertial pattern of the model and the slow response of real variables to monetary policy shocks, we posit an habit formation mechanism, based on the setup proposed by Campbell and Cochrane (1998), where households make their consumption decision on the basis of a single aggregate basket of differentiated goods. In addition to this, we also plan to study a second version of the model where the habit formation mechanism is related to a specific final goods variety, following the ‘deep habits' approach outlined by Ravn, Schmitt-Grohè and Uribe (2004). The advantage of such formulation is given by the higher degree of persistence of the demand of a specific final goods variety, which makes the markup countercyclical, according to the empirical evidence collected by Rotemberg and Woodford (1991). To make the interactions between monetary and fiscal policy more interesting, among the crucial assumptions of the model we plan to insert a proportional and a progressive tax system of production factor income. In fact, as discussed by Stockman (2001), in dynamic stochastic models there are remarkable welfare effects induced by proportional taxation system. Therefore, we are going to ask to what extent automatic stabilizers – modeled according to proportional taxation system – may affect the definition of the optimal monetary rule. With such analytical context we aim to investigate what is the optimal monetary rule, given a predetermined fiscal policy rule. A second question pertains the role of policy reforms. An economy like that previously described is full of distortions both on the real and the nominal side. An important question arises when policy authorities decide to insert an higher degree of competition by reforming labor or final goods market, conditioned to a given policy configuration (both monetary and fiscal). Therefore, it is natural to ask which reform ensures the highest welfare gain ? Labor market or final goods market liberalization ? The gain obtained by such reforms in terms of welfare is bigger than what is possible to obtain by implementing a fiscal reform replacing a progressive taxation system with a flat rate tax on productive factors income ? For the analysis of optimal monetary policy and of welfare effects derived from policy reforms, the metric for welfare level is defined by the expected utility of representative agent approximated up to second order, conditional to the initial steady state (constant and/or perturbed). To this goal, in order to avoid the inconsistencies showed in welfare calculations, it is necessary to solve the model by using a second order approximation, as suggested by Kim (2003). (ii) The second research topic pertains to the role of expectations heterogeneity. In this case, our purpose consists in presenting some important extensions to the contributions of the existing literature, by inserting a fiscal policy rule of the type described by FTPL. At the same time, by linking to the Research Unit of Bocconi University, we plan to examine the role of expectation heterogeneity of private sector and Central Bank in a simple model including also the term structure of interest rates. It will be possible to determine how the interactions between expectations of agents without rational expectations allows to determine equilibria which are expectationally stable (or E-stable) in the sense described by Evans and Honkapohja (2001). We assume that agents in the model do not have rational expectations, but they can learn about the nature of equilibria by using Recursive Least Squares algorithm to update their beliefs. We will study learnable equilibria with fiscal policy and expectation heterogeneity induced by a variety of monetary policy rules, as variants of the simple Taylor (1993) rule, by emphasizing the rules including expected inflation and output targeting. In particular, we plan to examine such results in the case where private sector and Central Bank are characterized by the same ability in forecasting future inflation and in the case where private sector agents formulate their expectations about future variables, by using the Central Bank's forecasts. In this case, agents – Central Bank and private sector – are characterized by different abilities in forming their expectations. The goal consists in examining: (i) the size of the noise produced by the type of agents with lower forecasting ability; (ii) the E-stability conditions with an explicit design of fiscal and monetary policies.(iii) For what concerns the third research topic, we plan to develop both on the empirical and the theoretical plan some specific hypothesis which can provide an interpretation of the evolution of the practice of government and corporate bonds indexation phenomena in European countries. In fact, it is easy to understand that for countries with monetary and fiscal authorities scarcely credible, bond indexation – especially government bonds – does not reduce the country default risk, since it is always possible the default option. On the other hand, countries with credible fiscal and monetary authorities (European Union Members), the indexation to a low inflation rate, presents considerable advantages both for the issuer and for creditors. For debtors (the issuers of bonds) the indexation to Euro inflation rate, reduces the effects produced by asymmetric shocks on real interest rate paid by the issuing Government. Therefore, bond indexation can turn out to be more or less cumbersome for the issuing country, depending from the type of shock hitting the economy. For creditors, the anchoring to Euro-based inflation rate makes more attracting the bonds for international investors; therefore, higher is the international diffusion of indexed securities among investors, more attractive is their indexation anchored to a widely recognized inflation rate. In order to develop such hypothesis, we plan to use several variants of traditional indexation models (as, for example, Fischer, 1975) to examine the role played by monetary and fiscal policies. In the traditional modeling framework the inflation rate is fully exogenous. Our goal consists in showing how bond indexation changes according to indicators defining fiscal and monetary policy stance, which affects the inflation rate according to the FTPL. This is going to be obtained via an explicit consideration of the government budget constraint in a model for the determination of the securities demand in continuous time. At the same time, we plan to construct a database aiming to understand the current practice of indexation in Europe and in Accession countries, trying also to link the indexation condition with a liquidity investigation of the bond market. EXPECTED RESULTS We can distinguish the expected results for the three research topics discussed previously.(i) For what concerns the first research topic, our goal consists in verifying with the theoretical model above described if optimal monetary policy rules show similar coefficients to those obtained via empirical estimates both for European Union and United States. In particular, we will study the conditions under which the output targeting coefficient of the monetary policy rules is negative or not, clarifying the puzzle identified by Schmitt Grohè and Uribe (2004). We plan to analyze such model by calibrating the various version to be presented on the basis of data for both European Union and United States. It is also worth to mention that the analytical framework to be developed can be employed to study a more general approach to the definition of the optimal monetary policy rule. However, our research will not be focused on a mechanical exam of Taylor rules within a new analytical context. In fact, we plan to verify if the introduction of different arguments in the monetary policy reaction function delivers optimal monetary policy rules with parameter configurations different from those considered in the empirical literature. This goal can be realized by considering inflation indexes computed by using the growth rate of nominal wages. Moreover, we are going to use the same framework to establish if the presence of a monetary aggregate in the monetary policy reaction function is welfare Pareto improving, given the emphasis considered by European Central Bank on monetary aggregates.(ii) For what concerns the second topic of the project, we plan to verify if the presence of fiscal policy widens the range of monetary policy reaction function parameter values which allow to define expectationally stable equilibrium. Moreover, if agents have different abilities in forming expectations , we will study to what extent agents with lower forecasting ability can affect also those with higher ability. This will imply a new definition of monetary policy parameter ranges previously considered in the literature in order to have a full coordination of expectations around the Rational Expectation Equilibrium. (iii) In order to develop the third research topic, we plan to build a model where the indexation dynamics is a function of fiscal and monetary policy stance. In this case, it will appear evident how the degree of bonds indexation evolves as a function of the commitment showed by the government in the stabilization of debt/GDP ratio. We also plan to develop an empirical model in order to show the evolution of indexation practice among European Countries and the evolution of the issuance costs (in terms of financial expenditures) characterizing the European Bonds Market, after the introduction of the Euro. In this context, we will assembly a database including also informations for the Accession Countries, i.e. Eastern European Countries which are candidate for the full membership of the European Union. For all the topics above considered, we plan to realize a series of working paper including the results of the research to be submitted to international journals and to be presented to international conferences.RESEARCH STEPSWe plan to articulate the research in two distinct steps.First stepWe plan to build and solve a theoretical dynamic stochastic general equilibrium model to be employed as a benchmark for the analysis previously highlighted. Among the main ingredients of such model will be habit formation. Progressive and /or proportional taxation. In this context, we plan to study the optimal monetary policy rule, conditioned to the fiscal policy rule, adopting the metric above discussed.For what concerns the topics on expectation heterogeneity, we plan to study to what extent fiscal policy rules proposed by FTPL change the monetary policy parameter configurations leading to an expectationally stable equilibrium. This type of analysis will be also studied for models including a specification for the term structure of interest rates. Finally, for what concerns the third topic, during this stage we plan to define a theoretical model to study how the degree of bond indexation changes according to specific indicators defining the fiscal and monetary policy stance. During this stage we plan to start the building up of the database to be employed in the empirical verification of the hypothesis discussed in the research program.Second StageDuring this second stage, we plan to extend the model studied in the first stage to the case of deep habits, where the markup becomes countercyclical. Even in this case, we plan to examine the optimal monetary policy rule. At the same time, we are going to present a comparative analysis of policy reforms under the form of labor and/or final goods market liberalization, as well as fiscal policy reforms. If we are going fulfill such goals in advance, we plan to extend out modeling framework to the open economy case, by working together with the Unit of Roma Tre.Moreover, in this second stage, we plan to extend the literature on expectations heterogeneity to a case including two policy authorities with a different expectation formation and a different information set, in order to verify if the monetary rules à la Taylor (conditional to fiscal rules of the type outlined by FTPL), allow to obtain an E-stable equilibrium, for a large variety of empirically plausible parameters.Finally, for what concern the indexation role, we plan to present some empirical work based on government bonds data, to provide a deep analysis of the evolution of the indexation practice, as well as to study the liquidity structure of the European Bonds Market.
|
