Unemployment and economic growth: a partial survey

Abstract

Summary The article surveys theoretical and empirical knowledge on the determinants of equilibrium unemployment. It is viewed as the outcome of the interaction between wage-setting and price-setting behaviour. The evidence suggests that generous unemployment insurance, product market regulations, high unionisation and uncoordinated bargaining at the sectoral level contribute to high unemployment. The evidence on active labour-market policy and taxes is less clear. But there is a strong presumption that job broking activities and labourmarket training reduce unemployment. The same holds for tax cuts on labour income only. It does not seem as if a high level of employment protection has an effect on the overall unemployment rate, although turnover in the labour market is reduced. Working-time reductions do not appear to raise employment. Much evidence suggests that also macroeconomic shocks can have long-term effects on unemployment. In an open economy, changes in product demand, which drive a wedge between real product and real consumption wages, can have long-lasting effects. Macroeconomic shocks can also interact with labour-market institutions. The relationship between long-term growth and unemployment is unclear. A higher growth rate can have both positive and negative unemployment effects. A key problem for labour-market reform is that it may not be in the interest of the political majority of employed insiders with medium and low incomes, as the main benefits may accrue to the unemployed, the high-skilled and shareholders. Status-quo bias and time inconsistency are other obstacles to reform. Political acceptance of reform may require compensating transfers to the majority that would otherwise lose out. * Lars Calmfors is professor of international economics at Stockholm University and Chairman of the Economic Council of Sweden. Bertil Holmlund is professor of economics at Uppsala University.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 64 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.