A workshop on Social Inequalities and Social Mobility was held in La Coruña, Spain, on 3/4 April 1998. The workshop, which formed part of a CEPR research programme on rising inequalities, was organised jointly with, and held at, the Instituto de Estudios Económicos de Galicia, Pedro Barrié de la Maza. The organizers were Andrea Ichino (European University Institute, Firenze, IGIER, Università Bocconi, Milano, and CEPR) and Gilles Saint-Paul (Universitat Pompeu Fabra, Barcelona, and CEPR).

Three papers focused on a range of issues relating to social and income mobility. Peter Gottschalk and Enrico Spolaore (both Boston College) presented 'On the Evaluation of Economic Mobility'. Trying to answer the question, 'Why do we care about mobility?', they considered preferences for those fundamentals – uncertainty, fluctuations in consumption – affected by the different aspects of mobility and provided a tool for assessing gains and losses associated with given transition matrices. They also considered the value of late versus early resolution of uncertainty, thus addressing the 'origin dependence' aspect of economic mobility.

In their paper, 'IQ, Social Mobility and Growth', John Hassler (Institute for International Economic Studies, Stockholm and CEPR) and José V Rodríguez Mora (Universitat Pompeu Fabra, Barcelona) investigated the links between social mobility and growth. They proposed an interpretation of social mobility as an endogenous phenomenon in a multiple-equilibrium model. When the market allows talented individuals, independently of their background, to become entrepreneurs, further knowledge and growth are developed, thus raising the return to individual talents for the gifted and raising wages for the non-gifted. The outcome is a high-mobility/high-growth equilibrium in which agents are sorted across occupations according to their built-in talents, rather than their social backgrounds. In the authors' interpretation, however, societies in which 'family' plays a decisive role in determining individuals' occupations will, in turn, experience little growth, thereby reinforcing the problem of low mobility.

Javier Ortega (Universitat Pompeu Fabra, Barcelona) discussed efficiency issues related to migration, or geographical mobility, in a paper entitled 'How (Good) Immigration Is: A Matching Analysis'. Starting from the lack of clarity about the effects of migration on natives' welfare, Ortega proposes a multiple-equilibrium two-countries matching model of the labour market in which agents can decide whether to search for work in their home country, or pay a migration cost and go abroad. On the firms' side, if migration is anticipated, more vacancies will be posted since lower wages will be paid to immigrants, who are high search-cost workers. This mechanism generates self-fulfilling expectations. The three equilibria found are also Pareto-rankable, and Ortega shows that the full-migration equilibrium is the most efficient, thus implying that natives do not suffer a loss from migration, and providing a rationale for several empirical results.

In their paper 'The Long-Run Educational Cost of World War II: An Example of Local Average Treatment Effect Estimation', Andrea Ichino (European University Institute, Firenze, IGIER, Università Bocconi, Milano, and CEPR) and Rudolf Winter-Ebmer (Universität Linz and CEPR), provide an assessment of the educational losses occasioned by World War II in several participant and non-participant countries. The estimation technique they employ – local average treatment effect – allows them to isolate the return to education for those persons whose educational choice was directly affected by the war in the sense that they were involuntarily pushed into a lower educational category. They estimate that in Germany in 1986, GDP was 0.36% lower than it would otherwise have been; the corresponding figure for Austria in 1983 was 0.67%. Ichino and Winter-Ebmer propose this measure as a proxy for the loss of human capital incurred because of liquidity constraints of importance comparable with those likely to arise during a war.

On the issue of inequality, in 'Income Redistribution Within the Life Cycle Versus Between Individuals: Empirical Evidence Using Swedish Panel Data', Anders Bjorklund (SOFI, Stockholms Universitet) and Márten Palme (Stockholm School of Economics, Stockholms Universitet) presented an analysis of the effects of the Swedish welfare state on inequality across individuals as well as within the life cycle. The authors used the generalized entropy measure to decompose overall inequality into these two components which were then computed for both pre- and post-tax and pre- and post-benefit incomes. They showed that taxes tended to reduce inequality across individuals, whereas the main effect of benefits was to smooth the path of income over the life cycle – although benefits also played a role in equalizing individual incomes. The income-smoothing effect of the welfare state was stronger in the lowest quartile of the long-run income distribution.

The paper entitled 'Wage Inequality in Spain and Portugal: Contrasts and Similarities', written jointly by Olga Canto (European University Institute, Firenze), Ana Rute Cardoso (Universidade do Minho) and Juan Jimeno (Universidad de Alcala and FEDEA, Madrid), presented evidence on the link between wage flexibility and earnings inequality. Their study looked at Spain and Portugal, two countries that share many institutional features and have experienced similar changes in labour demand and supply. Portugal's labour market, however, is regarded as one of the most flexible in the OECD, while Spain's is considered among the most rigid. Nonetheless, the results of the analysis suggest that there are fewer differences than might be expected between the earnings distributions in the two countries, although differences were found in the determinants of the distributions. Thus the reward for tenure was higher in Spain, whereas the return to schooling was higher in Portugal. Similarly, both skills and different types of collective bargaining agreements played more important roles in determining inequality in Portugal than in Spain.

Comparing France and the United States, Etienne Wasmer (Institute for International Economic Studies, Stockholm, and Centre for Economic Performance, LSE) offered a labour-supply interpretation of the rise in the return to skills. In 'The Labour Market Consequences of Demographic Trends: US and France 1964–94', Wasmer investigated the extent to which the substitutability of experience for education in the production function – which he quantified – affected the wage distributions and the returns to skills in the two countries. His conclusion was that changes in labour supply, if properly handled, could account in large part for most of the differences between the two countries.

In 'Equilibrium Unemployment Insurance', John Hassler (Institute for International Economic Studies, Stockholm, and CEPR), José V Rodriguez Mora (Universitat Pompeu Fabra, Barcelona), Kjetil Storesletten (Institute for International Economic Studies, Stockholm, and CEPR) and Fabrizio Zilibotti (Institute for International Economic Studies, Stockholm, and CEPR), presented a dynamic politico-economic model offering a rationale for the huge differences in unemployment benefits and rates in the United States and Europe. Human capital accumulation through learning-by-doing leads workers to vote in favour of high unemployment benefits since they would suffer significant losses following an unemployment-induced change of sector. Hence, if specialized workers are politically influential, society will agree to a high replacement ratio. This, in turn, implies longer unemployment spells and a more narrowly specialized and 'choosey' workforce. The opposite result arises if the initial workforce is less specialized: lower unemployment insurance payments will prevail, making the duration of unemployment spells shorter and workers more willing to change sector, thus reducing the benefits of learning-by-doing.