On average, tariffs are lower in countries with proportional electoral systems than in countries with majoritarian electoral systems. Using data on trade, institutions, and social policies in the rich democracies from the inter-war period to the 1990s, I argue that this well-known empirical regularity is best explained by the fact that proportional electoral systems produce more generous social policies, compensating losers from trade and lessening opposition to trade liberalization; the alternative explanation that majoritarian systems are associated with higher tariffs since legislators elected in single-member districts are more sensitive to protectionist pressures is shown to be of secondary importance.

The political economy literature has gathered compelling evidence that labour market risks shape political preferences. Accordingly, insecurity fuels support for redistribution and left parties. The paper analyses this argument for temporary workers, a so far neglected risk category which has increased dramatically in the past two decades. Temporary workers also have been in the focus of recent insider-outsider debates. Some authors in this line of research have argued that temporary work leads to political disenchantment, e.g. non-instrumental responses such as vote abstention or protest voting. This contradicts risk-based explanations of political preferences. The article discusses both theoretical perspectives and derives conflicting hypotheses for the empirical analysis of temporary workers\' policy and party preferences. The review reveals considerable ambiguity regarding the questions which parties temporary workers can be expected to support and what the underlying motives for party choice are. Synthesising arguments from both perspectives, the article proposes an alternative argument according to which temporary workers are expected to support the \'new\' left, i.e. green and other left-libertarian parties. It is argued that this party family combines redistributive policies with outsider-friendly policy design. Using individual-level data from the European Social Survey for 15 European countries, the article supports this argument by showing that temporary compared to permanent workers exhibit higher demand for redistribution and stronger support for the new left. Neither the risk-based nor the insider-outsider explanations receive full support. In particular, no signs of political disenchantment of temporary workers can be found. Thus, the findings challenge central claims of the insider-outsider literature.