RT Dissertation/Thesis
T1 Factor mobility and heterogeneous labour in computable general equilibrium modelling
A1 Flaig,Dorothee
WP 2014/06/27
AB The representation of labour markets in Computable General Equilibrium (CGE) models is characterised by a trade-off between data representation and data availability. Models are by definition abstract and simplified pictures of the real world: as a map of scale 1:1 does not help to find an unknown destination, a model which perfectly depicts the real world would hardly help to analyse adjustment effects of policy changes or macroeconomic shocks. When the analysis is focused on distributional issues, it seems obvious that such an analysis can only be based on models that differentiate at least more than one household group. Household groups characteristically differ in factor endowment and since factor income– besides price effects – is a main determinant of welfare analysis, the specification of labour markets crucially determines the analysis. There are mainly two possibilities to specify the labour market in a CGE model: First, the labour market can be set up as competitive market with perfect substitutability between individual workers on that market. With this setup, wages must be equal among labour types and sectors because every difference in wages provokes adjustments, which finally equalise wages again. In contrast, data reports typically significant wage differences between labour types that can only originate from imperfect labour markets. Thus, the second option is to depict these wage differences by imperfect substitutability of individual workers in the production process. But data on substitution possibilities of labour demand between different labour types is weak and estimations of substitution elasticities are in most of the cases not available.
Meanwhile, in the real world, wages differ in various dimensions and in models labour types are typically differentiated by age, gender, skill level or occupation. When differentiating labour types within these dimensions, wage differences become possible and can be explained by transformation limitations between characteristics: e.g., wage differences between female and male workers are originating from the fact that female workers cannot become male workers. This differentiation has the effect that in most of the models, transformation between the characteristics of a dimension is no longer possible and workers stay in a specific labour type. Typically labour types are not differentiated by sector of employment and, thus, are assumed homogeneous amongst sectors. Movement of workers between sectors seems possible; nevertheless, data reports partly huge wage differences between different sectors of an economy. As a solution, CGE models typically include an efficiency parameter which allows calibrating the model according to the data, but the model assumes still homogeneous labour which should be priced equal. Thus, the efficiency parameter does not economically explain the existence of these wage differences.
This thesis presents a comprehensive and flexible framework to introduce imperfect factor markets in CGE models. Labour mobility between labour types is controlled by migration functions where the degree of mobility is controlled by elasticities that govern the responsiveness of migration to changes in relative wages. Finally, the model provides the user with three additional instruments to control the operation of labour markets. First, the user can control the stock flow relationship for each labour type, e.g., does a migrating worker keep her productivity from the initial activity, adopt that of the destination activity or something in between; second, the user controls the flexibility of the labour market by setting the migration elasticities between activity blocks; and third, the setting of adjustment parameters determines the (assumed) costs of migrating. The analysis of productivity effects and costs of factor reallocation emphasises the relevance and influence of labour market specifications on model outcomes. Thus, this thesis sets the base for a careful setup and test of labour market assumptions applied in CGE models.
K1 allgemeines Gleichgewichtsmodell
K1 Arbeitsmarkt
K1 Faktormobilität
K1 Israel
K1 Palästina
PP Hohenheim
PB Kommunikations-, Informations- und Medienzentrum der Universität Hohenheim
UL http://opus.uni-hohenheim.de/volltexte/2014/982