Interrelationships between income and both religious and secular ethics: panel Data Envelopment and Stochastic Frontier Analyses combined

The focus on either ante or post secularisations and on either religious or secular ethics precludes the evaluation of their relative importance in income production. This paper suggests a theoretical model to empirically summarise the literature on the causal long-run interrelationships between Gross Domestic Product and both religious and secular ethics and to monetarily evaluate to what extent religious and secular ethics as well as changes in religious and secular ethics are inputs and desirable or undesirable outputs in producing Gross Domestic Product. To do so, it constructs a dataset of 152 countries over 26 years, by referring to 8 religions and 3 education levels and by distinguishing Developed and Former Soviet Countries. Next, it combines Data Envelopment Analysis and Stochastic Frontier Analysis, by estimating 3 empirical models. Stochastic Frontier Analysis replicates the main empirical insights from the literature (i.e., positive impact of Protestant Christianity, negative impact of tertiary education), by identifying two groups of countries characterised by different institutional contexts (i.e., Developed vs. Former Soviet Countries). Data Envelopment Analysis measures religious and secular ethics as well as changes in religious and secular ethics in terms of Gross Domestic Product (e.g., religious ethics at 5–7 %, secular ethics at 4–8 %), by showing that changes in religious ethics can be characterised as post and ante secularisation in Developed and Former Soviet Countries, respectively; changes in secular ethics favour a larger Gross Domestic Product to a greater extent in Former Soviet Countries; and a larger Gross Domestic Product favours changes in secular ethics to a greater extent in Developed Countries. In conclusion, this paper consistently applies Stochastic Frontier and Data Envelopment Analyses to find causal long-run bidirectional relationships of Gross Domestic Product with both religious and secular ethics, as well as to monetarily evaluate possible substitutions and transformations between religious and secular ethics. • One theoretical and three empirical models for income vs. ethics interrelationships. • A dataset of 152 countries over 21 years with 8 religions and 3 educations levels. • SFA and DEA applied to all, Developed and Former Soviet Countries (DC, FSC). • SFA repeats the empirical insights from the literature (e.g., Protestant, tertiary). • DEA highlights changes in religions as post and ante secularisation in DC and FSC.