Gerzensee Conference: Abstracts of Papers
Morning: Session I - An Economic Integration Session
Scenarios for Evaluating the Impact of Eastern Enlargement of the EU, by Maurizio Grassini.
ABSTRACT: This paper investigates the impacts of the proposed enlargement of
the EU. The group of countries being considered is heterogeneous. The
front-runners are Poland, Hungary, the Czech Republic, Estonia and Slovenia. These
are called the CEEC5. In the paper, the enlargement is effectively viewed as the
merging of two "countries", being the EU-15 and the CEEC5. The effect of
the enlargement upon the EU-15 is considered. Since the CEEC5 GDP is only 3.3% of
the EU-15 GDP, the enlargement is expected to have a negligible impact. Impacts on
Italy of the CEEC5 enlargement are also considered. Especially relevant is the
examination of the import-export structure of Italy vis-a-vis the CEEC5 countries.
Results of three scenarios are reported. The first examines the interactions between Italy and the CEEC5. The second examines the interactions between the EU-15 as a whole and the CEEC5. Finally, the third scenario examines specialization which may be expected to occur in the trade patterns of the CEEC5. Some of this specialization is already evident from the historical data. At the end of the paper, a continuation of the research programme is sketched out.
FUR and European Integration, by Paul Salmon.
The Economic Influence of a Japan-Korea Free Trade Area, by Toshiaki Hasegawa.
Afternoon: Session II - Inforum Problems and Improvements:
Some Reflections on the Empirical Foundations of Inforum Models, by Josef Richter.
ABSTRACT: Economic reality is primarily perceived through the eyes of the
statistical system of a specific country and, in particular, through the lenses of
input-output tables and national accounts. Inforum models attempt to forecast
economic development, or - more precisely - economic data that is eventually published a
few periods hench by the statistical agency. In the case of evaluating policy
alternatives, the aim is to simulate the statistical pictures that might occur under
The basis of statistical information, which is the empirical background of any Inforum model, contains elements that belong in three different categories of information:
1. Statistical data rooted in observations and which results from classifying and aggregating basic information.
2. Statistical data estimated by various types of models, although the target variables could have been observed - at least in principle.
3. Statistical data estimated on the basis of observed variables, although the target variable itself is not observable.
In the standard case, all three types of information are present and hardly separable. The implications of using data of different cognitive character are quite different depending on whether this data is used for testing purposes in the course of model building, or whether the data is used for forecasting or simulation purposes.
ABSTRACT: The fundamental input-output identities that express the solution for commodity output and price also guarantee that nominal GDP is the same whether measured as the sum of final demand expenditures or as the sum of value added. In the simplified framework in which the identities are usually presented, all relationships hold exactly, and there is one homogenous price for each commodity. However, when constructing a model of a real economy, one usually needs to confront several issues. First, prices may vary across the row. Second, input-output identities often do not hold with available data . Finally, final demand and value added from the input-output table may not be equal to the corresponding figures from the aggregate national accounts.
The solution to the first problem is simply to use a more complicated input-output identity. Often, the solution to the last two problems is to introduce discrepancy variables that can be added to final demand, unit value added, or aggregate final demand and value added concepts to force the necessary identities to hold. Such discrepancies are then usually held constant for the forecast interval of the model. This paper first discusses how non-homogenous prices are handled in several Inforum models. Next, we investigate the conditions under which nominal GDP as measured by total final demand or total value added is still equivalent, in the presence of final demand and value added discrepancies. Finally, several other types of discrepancies are discussed, that arise from inconsistent input-output and national accounts data.
Macroeconomic and Structural Effects of the German Environmental Tax Reform, by Bernd Meyer.
Experience with Optimization in Inforum Models, by Clopper Almon.
ABSTRACT: This paper describes how the technique of optimization has been tried in conjunction with estimating and running the quarterly QUEST model of the U.S. Optimization may be defined as the specification of an objective function that is defined over several estimated parameters of the model. Such an objective function may be the fit of levels or growth rates key variables in an historical simulation, or the satisfaction of some social goal, such as reduction of the "misery index". The first step in optimizing is to specify such an objective function, and pick the parameters of the model which should be changed to optimize the function. Then, the model is run in G7 with this optimization in effect. The paper presents results of improved historical simulations, as well as policy suggestions as to how the "misery index" could have been reduced over selected time periods of the 80s and 90s.
Morning: Session III, A Tax Effects Session
ABSTRACT: A first model-based analysis of the environmental fiscal reform in Germany has been undertaken using the econometric model PANTA RHEI and the general equilibrium model LEAN. The project was financed by the German Ministry of Finance and executed by a team of researchers of DIW, GWS (PANTA RHEI) and the University of Oldenburg (LEAN). The paper at hand presents the results, that were achieved with PANTA RHEI. The influence on economic growth is very low, employment is growing slightly, while energy consumption and CO2 emissions are decreasing. There is no structural change to the disadvantage of energy-intensive industries.
Stochastic Simulations with the Model PANTA RHEI, by Christian Lutz.
ABSTRACT: PANTA RHEI is a complex economy-energy-environment model
with 1173 estimated equations. In order to get a measure of the expected simulation
error of the model, stochastic simulations were performed. In multi-equation
simulation models, four types of errors are present:
1. Implicit additive error terms for each equation.
2. Estimated coefficients are themselves random variables.
3. Exogenous variables must be forecast, and these forecasts contain errors.
4. Equations may be misspecified.
In this exercise, the study was limited to the first type of error. Results are presented for the forecast of GDP and the unemployment rate, both as a band, and as a histogram. The conclusions are: 1) the effects of additive error terms in this model are not very important; 2) dynamic interdependences are small; and 3) we should also examine other sources of error.
ABSTRACT: At the 12th Inforum conference, a detailed SAM for Poland for the year 1995 was presented. In this paper, the SAM 1995 is used to simulate alternative policies on household and corporate sectors taxation. An influence of tax burdens on the whole economy's performance is taken under consideration. For each simulation, according to assumptions, the structure of the SAM is changed as well as the exogenous values. Thus, the comparison of the alternative variants concerns multipliers and the values of the main accounts.
Morning: Session IV - Labor Productivity, SWIMM Model
ABSTRACT: The paper reports some preliminary results on research we are undertaking in order to model the link between technical change, capital investment and labour productivity in the Italian model. In the first part a model of production that allows for both embodied and disembodied technical change is described and labour productivity and employment equations are derived from it. In the second part an estimation of the rates of the two components of technical change for the sectors of the Italian economy are presented. In the last part the results of the simulations of a version of the Italian model that embodies the new equations are reported.
New Productivity Equations for IMPEC: The Concept and First Results, by Michal Przybylinsky and Iwona Swieczewska.
ABSTRACT: A report presenting some findings while constructing new labor productivity equations for IMPEC. Changes in labor productivity are quite fast in Poland. The changes are the effects of both embodied and disembodied technical progress. The concept of the new equations is based on endogenizing Total Factor Productivity. The main sources of TFP growth are domestic R&D expenditures and technology transfer from abroad, represented by the FDI. This concept was verified by some preliminary estimations.
Value Added in SWIMM, by Samuel Guillet.
Afternoon: Session V
A Dynamic KLEM Factor Demand System for Dandy3m, by Peter Rormose Jensen.
ABSTRACT: The paper describes the development of a factor demand system for the Danish Inforum model DANDY3M. It includes the four production factors capital, labour, energy and materials. It is estimated as a Generalized Leontief system (GLO). The dynamic behaviour is captured by an error correction mechanism, which is estimated simultaneously with the long-run static equations.
The Analysis of Welfare Effects in INTIMO: A Preliminary Study, by Rossella Bardazzi.
ABSTRACT. The analysis of welfare effects on Italian economy after the EU enlargement to Eastern European countries is undertaken using the Italian Multisectoral Model INTIMO. The issue consists in finding an answer to the traditional microeconomic question: "what amount of compensation would the consumers require in order to forego the change brought by the enlargement?" A brief review of the theoretical and empirical literature on the topic is provided. Then, the feasibility of several approaches within the multisectoral model is analysed and a preliminary quantitative estimation of welfare changes is presented for the period 2000-2010.
Capital-Embodied Technological Change: Measurement and Productivity Effects, thesis of Dan Wilson, summarized by Clopper Almon.
A Note on Scaling, by Michael Field.
ABSTRACT: This note describes the operation of a new type of scaling, called "proportional scaling", which may be used in the revision or creation of economic data. Standard scaling is familiar to most who have worked on data and tried to make it sum to desired control totals. Right-direction scaling is a technique developed at Inforum to scale all elements of a group of numbers "in the right direction". Right direction scaling is described in Interdyme Report #3. Right direction scaling never changes the sign of the elements to be scaled. Proportional scaling may change the sign, under certain conditions.
Computer Demonstration: Visual Basic for Economists, by Frank Hohmann.
Morning: Session VI
A Regional Submodel for Andalusia: A Project for the Andalusian Regional Government, by Elena Alonso
ABSTRACT: The Andalusian regional government has asked Fundacion Tomillo to build a regional submodel of Andalusia which should be linked to the MIDE model. The purpose of the communication is to present our preliminary ideas to receive suggestions from other teams with experience in such type of projects.
The Role of Small and Medium Enterprises: Scenarios for Poland, by Mariusz Plich
ABSTRACT: Small and medium enterprises (SMEs) played a growing role in the Polish economy during the previous decade. They are estimated to account for 50% of the turnover in 1997 while the share was only 40% in 1993. SMEs, moreover, are of key importance for employment growth because of their flexibility and their ability to adopt better than larger firms to changing market conditions. In the paper some scenarios of Polish economic development are considered and their influence on the SME sector is analysed. A multisectoral approach is used in this analysis.
Effects of Koizumi's Structural Reform Plan on the Japanese Economy, by Takeshi Imagawa
ABSTRACT: The newly elected prime minister Jun-ichiro Koizumi has promised to the world to push his economic structural reform plan to revitalize the Japanese economy. His reform plan is comprehensive and consists of structural and regulatory reform including reduction of corporate debt and the banks non-performing loans. In this paper, we examine the effect of fiscal policy changes in his reform plan on the Japanese economy by performing policy simulations based on the JIDEA model.
Policy variables implementing the reform plan are the following:
1. Reduction of government consumption will be interpreted as the government administrative reform or reduction of government personnel.
2. Reallocation of government investment will be interpreted as the reduction of unnecessary public works projects.
3. Reduction in the issuance of government bonds for fiscal 2002 by a trillion Yen will be interpreted as the means to achieve a healthier and more balanced government budget.
The primary effects of the above simulation will be modified by reducing the number of government employees implementing the secondary or induced effects of the simulation. Finally, by calculating total factor productivity within the framework of the JIDEA model, the effects of Koizumi's reform plan on the Japanese economy will be evaluated.
Afternoon: Session VII
ABSTRACT: This presentation will be about the Ph.D. thesis of Dan Wilson, who used cross-section data across establishments to try to measure the rate of technological change embodied in capital. The application of this research in the U.S. Inforum model will also be described.