Centre party, district magnitude, and wage inequality in service economy

The post-industrial economy witnesses a dramatic shift from manufacturing to service sector employment. This transition has fueled significant debate about rising wage inequality, both between and within sectors. While much research focuses on the manufacturing-service gap, this paper addresses the often-overlooked wage disparity within the service sector itself. The service sector's growing dominance in employment and output necessitates understanding intra-industry inequality. The service sector's inherent heterogeneity – encompassing low-wage, part-time jobs alongside high-skilled, high-paying positions – complicates collective action by workers, making public policy intervention critical. Prior research emphasizes the importance of policies balancing flexibility and job security to foster growth while mitigating inequality. This paper argues that this crucial balance is more likely under centrist party governments in high-magnitude electoral systems. Centrist parties, due to their political independence from organized labor and industry, are better positioned to represent the unorganized interests of service workers. Furthermore, high-magnitude systems encourage these parties to form broader coalitions reflecting the diverse needs of the service sector workforce.

Existing literature on service sector wage inequality offers varied perspectives. Some categorize service sectors (dynamic, non-dynamic, welfare) and emphasize wage negotiation procedures and education in shaping inequality. Others highlight the need for policies balancing flexibility and security. However, the political economy of service sector wage inequality remains under-researched. While studies have explored the role of left-wing parties in mitigating inequality, their impact on the diverse service sector is ambiguous, with some arguing for pro-insider biases. This paper addresses this gap by focusing on the role of centrist parties and electoral system characteristics in shaping policy and outcomes.

The study employs a time-series cross-national dataset encompassing 14 OECD countries from 1980 to 2007. The dependent variable is service sector wage inequality, measured using Gini coefficients calculated across three service sector categories (dynamic, non-dynamic, and welfare services) based on Wren (2013)'s classification. Explanatory variables include district magnitude (Mean District Magnitude from IDB) and government partisanship (cabinet share of centrist, leftist, and rightist parties from Swank 2018). Control variables account for factors influencing wage inequality, such as centralization of wage bargaining, deindustrialization, GDP per capita, and age dependency. The analysis uses panel regression models including country fixed effects, between-effects, country and year fixed effects, and random effects to assess the impact of district magnitude, centrist party cabinet share, and their interaction on service sector wage inequality. Robustness checks include lagged partisan variables and the exclusion of the Netherlands to test for sensitivity to outliers.

Analysis of 10 OECD countries with proportional representation (PR) systems reveals a significant negative interaction effect between district magnitude and centrist party cabinet share on service sector wage inequality. Across various model specifications (country fixed effects, two-way fixed effects, random effects), increased centrist party representation in high-magnitude systems consistently correlates with lower wage inequality. This implies that centrist parties in such systems are more likely to implement policies balancing flexibility and security within the service sector. Visualizations confirm this interaction effect, with a clearer negative relationship in PR systems. In the full sample (14 OECD countries including 4 majoritarian systems), the interaction effect is weaker but still negatively correlated, though with less statistical significance. Between-effects analysis consistently shows a negative relationship between district magnitude and service sector wage inequality, suggesting that larger districts promote more egalitarian outcomes. Control variables yielded mixed results, with wage bargaining centralization positively impacting inequality, as expected, and GDP per capita showing a weak positive effect. Robustness checks with lagged partisan variables show results consistent with the main findings, and removing the Netherlands from the analysis does not negate the key interaction effect. The results suggest that the interaction effect is particularly pronounced in PR systems.

The findings support the hypothesis that centrist parties in high-magnitude electoral systems are more likely to pursue policies that balance flexibility and security in the service sector, leading to reduced wage inequality. This is because high-magnitude systems allow centrist parties to attract broader coalitions of voters including workers from diverse service sectors. The results highlight the significance of institutional design (electoral systems) and party politics in shaping labor market outcomes. The paper's contribution lies in demonstrating the crucial interaction between electoral systems and party politics, demonstrating that high district magnitude alone or a high share of centrist parties alone is insufficient to reduce inequality. Both elements are necessary for a balance to emerge. The weaker effects observed when including majoritarian systems suggests a context-specific effect, indicating that the observed relationship is stronger in PR systems with their inherent incentives for coalition-building.

This study demonstrates the significant interactive effect of district magnitude and centrist party representation in reducing service sector wage inequality, particularly within PR systems. Future research could explore the specific policy mechanisms through which this interaction operates, examining the content of labor market regulations and social welfare programs in countries with varying magnitudes and party configurations. Further research could also investigate the role of other political factors (e.g., union strength, interest group influence), potentially explaining the mixed effects in majoritarian systems, as well as cross-national variations in the service sectors themselves.

The study's time frame (1980-2007) limits its ability to capture more recent trends. The use of a specific dataset (EU-KLEMS) may restrict generalizability, while reliance on existing party classifications could influence the interpretation of partisan effects. Additionally, the analysis focuses primarily on OECD countries, potentially limiting its applicability to non-OECD contexts.