Publications and data generated by the MONROE project modelling toolkit Abstracts of working papers
Medium frequency cycles in non-frontier countries
In this paper we have introduced a Schumpeterian creative destruction model of the productivity dynamics of a generic non-frontier small open economy in dynamic stochastic general equilibrium. The integration of Schumpeterian growth and DSGE is not straightforward, but it opens the door to a more rigorous and complete analysis of medium frequency effects of shocks in non-frontier countries, than previously achievable only by DSGE or Schumpeterian growth models taken in isolation. The proposed model is a prototype, and therefore it purposefully abstracted from several important additional features that would increase the realism of the model and allow for reliable estimation and policy evaluations, but that would also introduce complexity to a level that would render the exposition much less transparent. However, our setting could be extended in a relatively straightforward way in important directions. Most notably are two. One is the explicit introduction of a government policy block, which would allow the simulation of government policies. Another is the introduction of sticky prices and wages, which would increase the realism of the model and allow for Bayesian estimation of all its parameters. These are certainly the directions for our next future research.
Market imperfections and institutional bottlenecks in incentivising R&I (PBL)
The aim of this working paper is to provide empirical and theoretical basis for the macro and regional economic modelling of the links between market structure and innovation as well as between institutional quality and innovation within the Horizon2020 MONROE project. Although there is a vast theoretical and empirical literature on the relationship between market structure and innovation, there has been much less attention on the link between innovation and institutional quality and even less on the combined effects of these two in the spatial inter-regional setting. Institutional quality in the paper is treated as a broad concept that captures law, individual rights and high quality government regulation and services. The goal of this working paper is two-fold: firstly, we would like to review the existing theoretical and empirical literature on the links between competition, institutional quality and innovation, and secondly, we would like to go a step further and propose a way to account for the links between market competition in space, the role of region-specific institutional quality and sectorial innovation in a context of a simple theoretical spatial computable general equilibrium (SCGE) model. This simple modelling framework provides the useful basis for the further development and refinement of our large-scale computable RHOMOLO-like SCGE model for EU at NUTS2 level within Horizon2020 MONROE project.
Productivity fluctuations in countries which only perform technology transfer rather than R&D (University of Surrey)
We built a DSGE model of technological catch-up in countries behind the global frontier. We go on to estimate it on Spanish and U.S. data, with the U.S. acting as the technological leader, and Spain as the follower. The model generates highly persistent movements in productivity in the emerging economy from un-persistent shocks to the technological leader, helping to explain Spanish data. Unlike the prior literature, our model would generate permanent productivity gaps between the global technological leader and other economies even were the leader’s technological progress to halt, helping to explain why the great recession has not been accompanied by convergence in productivity. This is due to the diminishing returns to technological transfer in our model. It is further exacerbated by competition between firms in the following economy, which reduce each firm’s share of production profits. We draw tentative conclusions on policies to enhance productivity in countries behind the global frontier.
Introducing endogeneity in the complementarity between information technology and skill biased technical change. (University of Surrey)
We provide evidence that the relative price and quantity of skill exhibit a very smooth medium term component in the US over the last 4 decades. This motivates the construction of a general equilibrium model in which endogenous growth interacts with skill and unskill-augmenting technical change and labor heterogeneity in production. When taken to the data this model produces a number of interesting results. It shows strong support in favor of the capital skill complementarity hypothesis and finds evidence of gross complementarity between capital/skilled labor and unskilled labor. Furthermore it shows evidence that both skill and unskill-augmenting technical change are important in explaining the evolution of the skill premium and none dominates the other. Finally it shows that a subsidy to research and development in an economy with capital skill complementarity could generate positive effects in the long run at the expenses of being contractionary on impact.
A new methodology to incorporate human capital and intangible assets in the CGE model
In this working paper a detailed description of how human capital can be introduced into the CGE framework is provided. The theoretical description of modelling human capital is accompanied by on-going empirical work. The relationship between labour productivity and education has been estimated using EUROSTAT data for all EU MS. In particular we estimated the labour productivity per hour worked across EU countries as a function of the share of low, medium and high skilled employees to total employment . Preliminary results support the argument that an increase in the share of the high skilled employees to total employment has a significant positive impact on labour productivity. The impact of different skills in labour productivity is more volatile in countries such as Bulgaria, Czech Republic, Hungary, Lithuania, Romania, and Slovakia, and less volatile in high-income countries such as Germany, France, UK and Italy. In addition the relationship between wage rate and education level has been estimated. The annual earnings by different education level from the latest structure of earnings survey 2014 have been used to estimate the relationship between the wage rate and the education level. Preliminary estimates showed a significant positive relationship between the wage rate and the education level. Workers at the highest education level have on average a 2.4 higher wage rate than the workers at the lowest education level. Another estimation that has been performed is on the relationship between the stock of knowledge and the multi-factor.
Introducing capital product and process innovation in a CGE model
We have developed a multi-sector model with different types of endogenous technical change: firm-level process innovation in the final sectors and product innovation carried out by manufacturers of investment goods, that is ISTC. Capital-embodied technical change is differentiated between general-purpose innovation, which decreases the relative price of investment goods as in classic ISTC models, and energy-saving innovation that makes new vintages of capital less energy-intensive. The model design fits the purpose of policy analysis, being possible to include inter-industry trade and Armington-style international trade. We propose a method that makes it possible to simplify the model implementation and calibration, especially when it is solved as a mixed complementarity problem. The advantage of this approach is to avoid imposing restrictive assumptions on the structure of the economy that, while delivering a BGP, would turn off key sources of structural transformation in policy simulations. In future research, the model and calibration method discussed here should be applied to specific policy questions, in particular those related to environmental innovation. While the focus in environmental policy is generally directed to environmental-friendly technologies, the use of our model might shed light on the role of more general innovation activities in achieving environmental goals.
Application of spatial heterogeneity in A DSGE model
Many important shocks in the real world are correlated not only in time, but also across some notion of space. This may be physical space, or the space of product, firm or household types. As a result of this spatial correlation, aggregate volatility emerges naturally from idiosyncratic shocks. In this paper, we introduce a tractable framework that allows for such shocks without necessitating the discretisation of space, or a departure from perturbation approximation. As a lead example, we construct a dynamic, stochastic, general equilibrium model of economic geography (DSGEEG). This model features population movement, firm dynamics and semi-endogenous growth. Using it, we show how transitory, spatially located productivity shocks can lead to permanent movements in population, helping to explain the decline of the U.S. mid-west. As an additional theoretical contribution, we derive conditions for the existence of continuous-in-space shock processes on a range of spaces of economic interest.
User manual for the use of the developed interactive online tool
The aim of this deliverable is to act as a manual for users of the MONROE Online Tool, which is a major deliverable of the MONROE project. The project has developed this interactive online tool, which, along with a modelling toolkit also developed by the project, give the opportunity to a broad range of stakeholders to evaluate the impacts of specific research and innovation (R&I) policies and programmes.
Read the full paper:
D7.3.2 User manual for the use of the developed interactive online tool
Technical description of the MONROE metamodel
The present deliverable describes the empirical setup, database and the results of econometric analysis that have been used as the basis for the construction of the meta-model used in the interactive part of the MONROE online tool. The goal of the MONROE online tool is to provide a broader audience of policy makers, students and other interested citizens information about the importance of R&D, the channels of impact and likely effects of sectoral economic growth of various R&D related policy measures.
Read the full paper:
D6.5.1 Technical description of the MONROE metamodel
Key messages and recommendations for policy makers from the results
This paper concentrates on the key messages and recommendations for policy makers resulting from the MONROE.
Read the full paper:
D6.4.2 Key messages and recommendations for policy makers from the results
Technical description of the modelling results, including interpretation of the differences between the outcomes of the models
One of the key aims of the EU has been to encourage increasing levels of research investment, in order to provide a stimulus to the EU’s competitiveness. The Europe 2020 strategy adopted in 2010 maintains a headline objective, namely, for the EU to devote 3% of its GDP to R&D activities. The EU strategy proposes a European economy based on knowledge, research and innovation (R&I) in an effort to create growth, quality jobs and help address societal challenges. Investments in R&D are a key component of the EU’s strategy that should be complemented with improved framework conditions and the supply of high labour skills and capital. However, Europe spends less on R&I activities relative to its major trading partners, USA, Japan and China, and almost all EU Member States have still not met their R&D intensity targets for 2020.
Comprehensive set of model results from the EU-EMS model
The main goal of deliverable ‘D6.3.5 Comprehensive set of model results from the EU-EMS model’ is to present the outcomes of policy simulations that use the R&I module of the EU-EMS modelling system. A full description of policy scenarios, the underpinning assumptions and the modelling channels for each policy type is given in deliverable D6.4.1.
Read the full paper:
D6.3.5 Comprehensive set of model results from the RHOMOLO model
Comprehensive set of model results from the PACE model
The PACE model is particularly appropriate to analyse macroeconomic and sector-level effects of EU climate and energy policies. Based on the extensions carried out within the MONROE project, specific R&I policies will complement this policy analysis. All policy scenarios assume public budget neutrality.
Similar to the scenarios designed for the E3ME model, the policy scenarios simulated with PACE have been constructed such that an additional policy is introduced on top of the central policy scenario S0 which is common to all models within the MONROE project.
Read the full paper:
D6.3.4 Comprehensive set of model results from the PACE model
Comprehensive set of model results from the DSGE model
Comprehensive set of model results from the E3ME model
See full results here:
D6.3.2 Comprehensive set of model results from the E3ME model
Comprehensive set of model results from the GEM-E3 model
The constant creation of new ideas and their transformation into technologies and products forms a powerful cornerstone for 21st century society, with many universities, research institutes and innovative companies cultivating this process. The EU strategy calls for the delivery of stronger, lasting growth and the creation of more and better jobs while respecting and promoting social and environmental objectives. The strategy proposes a European economy based on knowledge and innovation, ensuring that innovative ideas can be turned into new innovative products and services in an effort to create growth, quality jobs and help address societal challenges. Investments in Research & Innovation (R&I) are a key component of the EU’s strategy for the future of Europe. Research organisations, universities and technology and innovation-based EU companies require higher investment in R&I as well as better framework conditions in areas such as market regulation and the supply of skills and capital. Policymakers are interested in understanding how specific types of innovation affect firm productivity, economic growth and society.
Read the full paper:
D6.3.1 Comprehensive set of model results from the GEM-E3 model
Description of scenarios and output indicators that will be assessed in MONROE (CE)
Investments in Research & Innovation (R&I) are a key component of the EU’s strategy for the future of Europe, but – as emphasised in the report ‘New Horizons: Future Scenarios for Research and Innovation Policies in Europe’1 – strong universities, research organisations, laboratories and technology-based companies in the EU will require much more investment in R&I as well as better framework conditions in areas such as market regulation, skills and the supply of capital.
Policymakers are interested in understanding how specific types of innovation affect firm productivity and economic growth, yet many economic models currently used by the European Commission for the assessment of R&I policy are not specifically designed to deal with R&I. The aim of MONROE is to inform policy makers about the likely outcomes of R&I policies and programmes, through the successful integration of R&I in a set of macroeconomic models. The suite of models included in MONROE – the EU-EMS model (PBL), the PACE model (ZEW), the GEM-E3 model (E3M), E3ME (CE)and a newly created DSGE (UoS) – have each been set up to better assess the impact of R&I policy instruments.
Description of baseline case to be used in the model (CE)
The objective of MONROE is to analyse how research and innovation (R&I) and related policies affect the European economy, at macro, regional and sectoral level, with a set of models. The modelling in the project provides an assessment of how an intervention affects the economy through a scenario-based approach. Analysis of the impacts in the scenarios requires a comparison against what the outcomes would have been in the absence of intervention. This information is found in the baseline, and the differences in outcomes between the results in the baseline and the results in the scenarios are attributed to the policies that are assessed.
Read the full paper:
D6.2.1 Description of baseline case to be used in the models
Technical description of the R&I module of EU-EMS model
The main goal of this deliverable is to describe in detail the theoretical and empirical approach for the representation of impacts of R&I investments and R&I-related policies on R&I intensity and the development of regional and sectoral productivity in the EU-EMS modelling system of PBL. The focus of the current deliverable is on presenting the theoretical framework and corresponding empirical econometric analysis of R&I module in EU-EMS modelling system. It is not the aim of this deliverable to demonstrate how concrete policy measures and policy packages can be analyzed using this new module. The follow-up MONROE project deliverable ‘D6.3.5 Comprehensive set of model results from the EU-EMS model’ describes in details policy packages, modelling channels for each policy type and the outcomes of policy simulations that use R&I module of EU-EMS.
Read the full paper:
D4.5.2 Technical description of the R&I module of RHOMOLO model
Technical description of the R&I module of the E3ME model (CE)
Policy makers are interested in understanding how specific types of innovation affect firm productivity and economic growth, yet many economic models currently used to assess R&I policy are not specifically designed to deal with R&I. The aim of MONROE is to inform policy makers about the likely outcomes of R&I policies and programmes, by improving treatment of R&I features in a set of macroeconomic models, one of which is E3ME.
The outcomes of the policy scenarios, modelled by E3ME and the other models in MONROE, will feed into the interactive online tool that will allow a broad range of stakeholders such as European and national governments, academia, regional and local bodies to quantitatively evaluate the impacts of their specific R&I policies and programmes. Because some of the measures require new treatments within E3ME, this report presents the technical developments and modelling methodologies transparently.
Read the full paper:
D4.4.2 Technical description of the R&I module of E3ME model
Technical description of the R&I module of GEM-E3-RD model
This report provides a technical description and the algebra related to the model developments that took place within the MONROE project. The GEM-E3 model has been extended so as to include: five labour skills, endogenous skills supply, endogenous knowledge absorption and stochastic public R&I decisions. This report is structured in the following way: i) the next section provides the technical description regarding the skills differentiation and endogenous labour skill supply, ii) the third section provides the technical description regarding the endogenous knowledge absorption and iii) the last section focuses on the public R&I decision in the new GEM-E3 model.
Read the full paper:
D4.3.2 Technical description of the R&I module of GEM-E3-RD model
Technical description of the R&I module of PACE model
The quantitative assessment of endogenous R&D is carried out with PACE (Policy Analysis based on Computable Equilibrium), a multi-sector, multi-region computable general equilibrium (CGE) model of global production, consumption, trade and energy use that has been extended to include an endogenous R&D module. The implementation of PACE is done using MPSGE (Mathematical Programming System for General Equilibrium Analysis; Rutherford, 1999), which is a subsystem of GAMS (General Algebraic Modelling System; Brooke et al. 2010). The solver used in the analysis is PATH (Dirkse and Ferris 1995) for solving the MCP (mixed complementarity problem). PACE can run starting from 2010 until 2050 in five-year time steps and solves for a sequence of market equilibria. The global economy is described by a set of equations (i.e., market clearing, zero profit, and income balance conditions). For each year, the solution algorithm finds the set of prices and quantities that solves these equations.
Read the full paper:
D4.2.2 Technical description of the R&I module of PACE model
Documentation and results of estimating the new DSGE model with endogenous growth for Europe
This paper builds and estimates a multi-country DSGE model with endogenous growth for policy analysis. The model features multiple countries, multiple sectors and multiple policy levers. The model features an international technological frontier, advanced by incremental process innovation, based on the work of Holden (2016; 2017). Product innovation absorbs long-run scale effects, while firm entry absorbs much of the impact of short run fluctuations in demand on process innovation. Slow international diffusion of productivity improvements is driven by complementarities between human capital and advanced technologies. Intuitively, new human capital is needed to understand new technologies. The model is estimated on data from 1870-2013 for six regions. The estimated model suggests policy can have powerful effects on the level of output over the medium term, and even some permanent effects. However, these long-run effects are swamped by the model’s endogenous persistence even over forty-year horizons.
The role of human capital in creating knowledge, increasing productivity and absorbing spillovers (E3M)
The objective of the working paper (D 3.4.2) was to empirically estimate the relation among total factor productivity with human capital, R&D performance and the ability to generate and absorb knowledge spillovers so as to parameterize and increase the empirical relevance of the endogenous technical change module of the GEM-E3-RD model.
A literature review on the findings of current empirical estimations is provided. New econometric estimates performed within the MONROE project and were designed to match both recent empirical findings and with the mechanisms included in the model, GEM-E3-RD. The link between the econometric estimations and the GEM-E3-RD model is presented. An illustrative simulation of the properties of the model before and after its parameterization is made.
Open Source code implementing the approach as an add-in in Dynare
Working paper describing the approach for estimating models without balanced growth paths (Deutsche Bundesbank and University of Surrey )
This paper presents an approach for solving, simulating and estimating models without balanced growth. Key to the simulation method is the addition of an extra unit root variable which acts as a growth switch. We discuss ways of further improving the accuracy of this method, including showing how models may be approximated around a non-steady-state point. We also provide a second order extended Kalman filter for DSGE models solved at second order which tractably propagates a Gaussian approximation to the state distribution without approximating the mean or covariance. In an application, we construct a real business cycle with CES production and preferences, and estimate it on a long span of annual US data. We show that the model is able to explain the decline in the labour share over time
Working paper on Public R&D investment and the Labour Share (University of Surrey)
This paper studies the effects of public investment in R&D on the labour income
share in the US and in a pool of selected advanced economies. Empirical results
show that an increase in public investment in R&D produces a substantial decline
in the labour income share in the short run together with the standard long run
positive eects in terms of output, investment and consumption. This means that
public investment in R&D can generate non-negligible (re)distribution eects in the
short run that can potentially harm agents that are heavily dependant on labour
income. In the second part of the paper I try to rationalise this result through
the lens of a DSGE model with endogenous technology adoption. In the model a
government subsidy towards adoption of new technologies induces rms to substitute
resources away from capital accumulation and towards the R&D process. This initial
decline in capital accumulation generates a decline in hours worked consistent with
complementarity in production between capital and labour. This in turn drives
down labour income that, together with the increase in output, induces the fall in
the labour share.
Read the full paper:
D2.6.2 Public RD investment and the labour share
Working paper on modelling stochastic public R&I decisions including financial constraints
The aim of this report is to present the current status on public R&D modelling and suggest a methodology to endogenise public R&D decisions on clean energy technologies in the GEM-E3 model. The study focus on how the public sector currently spends its R&D budget on clean energy technologies, what is the learning potential for these technologies and how the public sector will make an optimal selection regarding R&D budget allocation on clean energy technologies under uncertainty. Data on R&D budget and learning by research rates have been extracted from the EUROSTAT, OECD and IEA databases and the respective literature.
Working paper introducing financial frictions into a model of endogenous growth and entry (University of St. Gallen)
The working paper “Introducing financial frictions into a model of endogenous growth and entry” has been accepted for publication by the Review of Economic Dynamics with the changed title “Innovation and Inequality in a Monetary Schumpeterian Model with Heterogeneous Households and Firms”.
Questions or remarks? Leave your question or remark down below, the MONROE project team will usually answer within twenty four hours on office days (Monday to Friday).