Public debt and economic growth : is there a threshold in Switzerland?

This thesis investigates whether a threshold exists in the relationship between public debt and economic growth at the cantonal level in Switzerland. Given the country’s federal structure and high degree of fiscal decentralization, the analysis is designed to capture the heterogeneous economic, demographic, and institutional characteristics across cantons. The central research question asks whether there is a debt-to-GDP threshold beyond which economic growth is negatively affected, and which regional variables shape this relationship. The theoretical framework builds on classical, Keynesian, and modern empirical literature that identifies possible linear, nonlinear (threshold), and asymmetric effects of public debt on growth. It also incorporates institutional factors such as the Swiss debt brake and the national fiscal equalization system, which play a crucial role in cantonal fiscal sustainability. Unlike many studies using national gross debt levels, this analysis applies net public debt and addresses nonstationarity by using Hodrick-Prescott (HP) filtering, focusing on cyclical deviations from long-term debt trends. The study uses panel data from all 26 Swiss cantons for the years 2009 to 2022, including variables such as human capital (proxied by education spending), a resource index representing fiscal capacity, population growth and commuter flows. Due to data limitations, the analysis clusters cantons into two groups, namely heavily urbanized and less urbanized, to identify regional patterns more effectively. Econometric techniques include threshold regression, asymmetric effect modelling, and Granger causality tests. While statistically significant threshold effects could not be confirmed, regime-specific coefficients suggest that negative growth effects intensify when debt levels rise above their long-run trend, particularly in less urbanized cantons. Asymmetric effects were clearly identified: debt increases significantly reduced growth, while debt reductions had no comparable positive effect. Granger causality tests revealed no predictive power from debt to growth but confirmed a strong predictive relationship from human capital to growth in both canton types. However, the negative coefficient on human capital suggests that short-term returns on education spending may be limited, reflecting delayed effects or spending inefficiencies. The findings imply that rigid debt thresholds may not be suitable in the Swiss-cantonal context. Instead, policymakers should adopt flexible, regionally tailored fiscal targets, prioritize the prevention of excessive debt accumulation, and focus on the efficiency and alignment of public spending, particularly in productive expenditures. For structurally weaker cantons, fostering economic diversification and enhancing labor market connectivity may improve resilience. The study underscores the importance of granular and context-specific fiscal analysis in shaping sustainable and growth-oriented policy in decentralized systems.