50th Congress of the European Regional Science Association: "Sustainable Regional Growth and Development in the Creative Knowledge Economy", 19-23 August 2010, Jönköping, Sweden
One of the key issues in economics is the explanation of unemployment. Doing so "modern mainstream macroeconomics" frequently refers to institutional structures in the individual countries (e.g. Layard, Nickell & Jackman 1991, 2006; Carlin & Soskice 2006). However, unemployment within states varies as much as between these countries. In Germany, for example, there are regions in which even during the crisis virtual full employment prevails (e.g. Munich Area) and others which face a deep labour market crisis (e.g. Ruhr Area). Within a country, however, there are only minor differences in the institutions. Therefore the large variation in regional unemployment is puzzling. We explain this regional variation of unemployment building on structural change and technical progress. The price elasticity of demand changes across the industry/product life cycle. The more mature an industry is the less elastic is demand. Technical progress has two opposing effects on employment. Increased productivity allows producing a given quantity wit less labour. This is the displacement effect of technical progress. However, technical progress decreases costs which lead to a drop in price. This in turn increases product demand and therefore labour demand rises. A compensation effect occurs. How strong this effect is and whether it may even "overcompensate" depends on the price elasticity of demand and therefore on the industry life cycle. However, each region has a specific industry mix and industries are regionally concentrated (Krugman 1991). The development of a region depends strongly on the demand elasticities of the dominant industries. Thus the mechanism of product life cycle, price elasticity and technical progress leads through the specialisation of regional economies to different spatial development paths of employment. We show formally that a transition from the elastic into the inelastic demand function for the dominant industries can plunge a region into crisis. Empirically we estimate Marshallian type demand functions using an instrumental variables estimator and industry level data to derive the price elasticities for different industries. In a second step these elasticities and the regional industry structure are used to explain regional (un)employment. The results support our theoretical reasoning.