This paper aims to statistically discover the relationships between decentralisation and happiness which has been continuously discussed up until now, to build a causal diagram enabling to understand the structure containing the independent, the mediating, and the dependent variables involved in it, to construct policy scenarios beneficial to policy-makers, and to forecast future conditions to which policy scenarios can lead under a certain circumstance. In doing so, the author employs System Dynamics approach which is a method to enhance learning in complex systems and also is grounded in the theory of nonlinear dynamics and feedback control. The variables included in the analysis are one dependent variable (level of happiness) and four independent variables (GDP per capita, level of regulation, level of decentralisation, and level of political democracy). The analysis results show that happiness is closely statistically related to four independent variables employed in this analysis, can be structurally understood by a causal diagram and also the policy scenarios regarding it can be forecast by using System Dynamics approach.