Abstract

In most of the literature on industrial relations, union members are found to be less satisfied with their jobs than non-members. Analysts have applied various statistical and econometric approaches to try to resolve what seems like a paradoxical finding, with mixed results, using theories based on selection bias and ‘exit-voice’ considerations. We review this literature, and note that most empirical studies are from only a few countries – especially the US and the UK. Analysis of a wider range of 18 countries participating in the large-scale European Social Survey in both 2006 and 2010 finds that trade union members generally tend to express higher rather than lower job satisfaction than others, although results differ by country. We use regression models (ordinal logistic) to show that union membership is generally associated with higher job satisfaction, even after controlling for individual, job and workplace differences. Attempts to link the union
factor in job satisfaction to typologies of countries, either by welfare regime or extent of
collective bargaining coverage, have not been able to address the puzzle of why a negative
link persists in a few countries, but not in most.