Does income matter in the happiness-corruption relationship?. Issue 3 (6th May 2014)

• Record Type: Journal Article
• Authors: Arvin, Mak; Lew, Byron
• Is Part Of: Journal of economic studies. Volume 41:Issue 3(2014)
• Page Start: 469
• Page End: 490
• Publication Date: 2014-05-06
• Subjects: Corruption -- Happiness -- Income thresholds; Economics -- Periodicals; 330.05
• DOI: 10.1108/JES-02-2013-0024
• Languages: English
• ISSNs: 0144-3585

Abstract

Abstract : Purpose: – Empirical evidence on the relation between happiness (life satisfaction) and corruption is barely perceptible in the literature. The purpose of this paper is to contribute to closing this gap by presenting some estimates using a large cross-section of countries over the period 1996-2010. Design/methodology/approach: – The empirical model allows both corruption and per capita income to enter as arguments of a happiness "production function". The correlation between happiness and corruption is presumed to be non-linear. Findings: – While the results do not support the existence of a Kuznets-type trajectory, the study finds that the level of per capita income determines whether happiness and corruption are related and in what way. The authors estimate cutoff income levels at which corruption has a discernible effect on happiness. The results show that corruption reduces happiness, but only for high-income countries – roughly the upper half of the income range in the sample. Practical implications: – Results nullify the oft-asserted statement that happiness is negatively linked to corruption in all countries. The nature of correlation is more complex. Originality/value: – The paper goes beyond simply testing whether happiness is related to corruption. It conjectures that the relationship between the two variables is non-monotonic. Thus, the analysis considers the notion that the association between happiness and probity is income dependent. A novel feature of the empirical model is that the estimated income cutoff levels are endogenously determined. That is, income thresholds are not pre-determined. The authors also test for the robustness of the results by addressing the issue of potential endogeneity of corruption.