RICERCA ITALIANA     

Research program

New Keynesian dynamics in unionised labour markets and the design of EU institutions

University Co-ordinator

Università degli Studi di ROMA "La Sapienza" - ECONOMIA PUBBLICA - ROMA(RM)

Research Unit Leader

Nicola ACOCELLA

Description

The research direction of the Rome Local Unit is twofold: on one side, we aim at developing schemes for the evaluation of the main economic institutions characterizing the EMU context; on the other side, the same analysis should be capable to test some consolidated claims offered by the recent macroeconomic analysis with respect to endogenous distortions and model uncertainty in policy games. Particular attention will be paid to the policy implications for the reform of the Stability and Growth Pact and for the possible developments of an informal policy coordination set-up such as the Cologne process. As a reduction of the fiscal burden may be ineffective if its impact on labor market performance is neglected, neo-corporatist institutions could prove to be valuable tools in enhancing macroeconomic performance.
More in detail, the program focuses on the interaction between fiscal and monetary policies in a New-Keynesian (monopolistic competition) framework under these basic assumptions: i) wage setters are not atomistic; ii) they strategically interact with both fiscal authorities and central bank; iii) information problems about the true model of the economy may exist.
The plan of the research can be articulated into four sequential steps.
In the first step, the aforementioned issues will be tackled using static models, so to better single out the nature and the consequences of the interaction among the different institutions under a "stationary" point of view. The main focus here is on the different institutions that can operate in a monetary union, such as the common central bank, governments, and labor unions. We will consider some alternative ways of devising each of them and analyze the consequences of their interactions under conditions of certainty and uncertainty. In particular we will look at "cooperative" institutions, as, by definition, a monetary union is a highly integrated economic area with a large number of interactions between the participating countries. Given the presence of externalities, the design of macroeconomic coordination is a crucial issue, but coordination is not a trivial outcome. For instance, one could argue that the introduction of a common monetary policy and restrictions on fiscal policy at the national level increase the need for macroeconomic policy cooperation due to the presence of externalities. However, given the potentially adverse reaction of the central bank, fiscal coordination could be counterproductive or ineffective. In other words, the introduction of additional players like an additional country, the central bank or national labor unions can have dramatic effects on the standard propositions based on two-country models, as e.g. Rogoff (1985a) or Kehoe (1988). On the operative side, this line of research will be pursued by merging the recent literature on non-atomistic wage setters and the Dixit and Lambertini's New Keynesian models, thus making labor market distortions endogenous to the fiscal-monetary interaction. This should provide us with a encompassing and motivated framework in which casting the investigation on the effects of fiscal coordination, neo-corporatism, social pacts, model uncertainty, and central bank's transparency.

The second step will be to shift the analysis from the stationary context to an explicit dynamic environment. This will be pursued by integrating the main element of the LLMI models, stemming form the seminal Barro and Gordon (1983) contribution, into the theoretical framework provided by the NK-DSGE analysis. As previously discussed, the LLMI strand of literature links the nature and the functioning of the labor markets (particularly the degree of wage distortion induced by unions in the bargaining process) to the structural components of the monetary stance, as determined by the degree of conservativeness in the central bank's objective function. This analysis offers an explanation of the macroeconomic performances of different institutional set-ups based on the matching of those two factors. Anyway, in most of the LLMI models (e.g. Cukierman and Lippi, 1999, 2002; Guzzo and Velasco, 1999, 2002; Cukierman Coricelli and Dalmazzo 2000, 2004) wages are predetermined but prices are fully flexible. The core element of the NK-DSGE analysis is the introduction of distortions in the price and/or wage determination process into dynamic stochastic general equilibrium models. Specifically, in the standard versions of the NK-DSGE framework, the mechanisms underlying prices adjustment through time are also the effective source of the models' dynamic features. The inclusion of price stickiness in a dynamic fashion is accomplished with the adoption of a Calvo-type rule for price adjustment or with a dynamic law of nominal price variations based on a "sticky information" mechanism (Mankiw and Reis 2002). In spite of their diffusion, some of the results of NK-DSGE models are still unsatisfactory and ask for further work. This is particularly true for topics strongly related with our research interests, such as the impact of fiscal policies and the effects of fiscal and monetary policy interactions (e.g. Woodford and Benigno, 2003). Thus, the proposed integration between LLMI and NK-DSGE seems to be natural and potentially fruitful to our aims.

The next phase of the research will deal with extensions of the aforementioned integration between LLMI and NK-DSGE analysis. Specifically, we will consider the role played by model uncertainty in the interaction between fiscal policy, monetary policy and unionized wage setting, paying particular attention to uncertainty on central bank's preferences (Beetsma and Jensen, 1998). From a theoretical perspective, considering information as a fundamental strategic variable in the hands of the central bank may have important implications for the evaluation/definition of policies and public institutions (Estrella and Mishkin, 1999; Rudebusch, 2001 and 2002; Walsh, 2003, Section 4.
Within a standard Barro-Gordon framework, Brainard's (1967) "conservatism principle" may improve social welfare by reducing the inflationary bias generated by a central banker's inflationary temptation, but the same may apply in the absence of an inflationary bias and in the presence of uncertainty on the central bank's preferences (for a survey, see Di Bartolomeo, Giuli and Manzo, 2005). But this "conservatism principle" has been recently challenged by a number of contributions fuelling passionate controversies. For instance, if extreme uncertainty about the shocks affecting the system is introduced in Ball's (1999) model, very aggressive policy rules turn out to be optimal (Sargent, 1999), whereas a focus on the real time data uncertainty leads to attenuations of the optimal policy rule (Rudebusch, 2001). Similarly, when the central bank is concerned with both output and price stabilization, uncertainty about the persistence of inflation may lead to policy rules more aggressive than under certainty equivalence (Söderström, 2002), whereas in the presence of stochastic fluctuations in the linkage between output gap and inflation multiplicative uncertainty leads to caution in a straightforward Brainard fashion (Walsh, 2003) but also increases the optimal weight that should be placed on the inflation target. In the face of this contrasting evidence, it is increasingly recognized that disagreement may come more from the different possible specifications of model uncertainty than from other sources (see Onatski and Williams, 2003).
The interest in the topic is also motivated by the observation that private information, together with the expectations' formation process (see, for example, Faust and Svensson, 2002; Honkapohia and Evans, 2001), represent key elements to understand the transmission mechanism of monetary policy (see, for example, Smets and Wouters, 2002a) and the interaction between monetary and fiscal policy (see, for example, Benigno and Woodford, 2003). Furthermore, from an applied perspective, the notion of model uncertainty is already fostering empirical analyses mainly in the field of monetary policy rules (Brock, Durlauf and West (2003); Levin and Williams (2003); Onatski and Williams, (2003).
After drawing on the literature on learning, expectations and the manipulation of information (see Honkapohia and Evans, 2001) we aim at building a dynamic New Keynesian model with the following features: i) it will display a stylized but enough comprehensive representation of the main characteristics of the European economy; ii) it will be capable to integrate in the analysis the role played by the Central Bank's information (and transparency); iii) it will allow, in particular, to explicitly investigate the impact of transparency in the expectations' formation process; iv) it should help in clarifying the role of private information in the fiscal-monetary policy interaction in the light of recent theoretical developments provided by the related literature (Benigno and Woodford, 2003). This analysis will allow us to address the problems of coordination arising between a Central Bank and multiple fiscal authorities in presence of substantial forms of uncertainty in the economy. Results to be drawn form the proposed analysis will help to shed light on the role played by preference opacity, vis-à-vis credibility, as a device to discipline wage setters in dynamic context. More in particular, we wish to understand how results change with the specification of fiscal policy, and in particular with its degree of activism in supporting employment, which in turn depends on the attention which is paid to (or the constraints which are placed upon) public deficits.

Finally, the last step will involve extension of the analysis to the case of an open economy characterized by a Phillips curve augmented with mixed backward and forward expectations, incomplete international financial markets and thus imperfect ‘risk sharing.'