Religiosity as a Determinant of Happiness

• Author: Erich Gundlach,Matthias Opfinger
• DOI: http://doi.org/10.1111/rode.12047
• Published date: 01 August 2013
• Date: 01 August 2013

Religiosity as a Determinant of Happiness

Erich Gundlach and Matthias Opfinger*

Abstract

The empirical relation between happiness and religiosity is considered from the perspective of basic utility

theory. An unbalanced cross-country panel data set is used to study whether religiosity can be considered

as a substitute in the happiness function, which itself is held to be a proxy for the utility function. We find

that the same level of happiness can be maintained with high and low levels of religiosity as a result of sub-

stitution along a standard indifference curve. Our empirical results are consistent with three stylized facts of

the empirical literature, namely a positive correlation between happiness and religiosity, a positive correla-

tion between happiness and income, and a negative correlation between religiosity and income.

1. Religiosity, Happiness, and Utility Theory

Religious behavior does not generate a direct financial reward to believers, but it

cannot be questioned that religious activities generate subjective happiness. Hence

religiosity appears to be a plausible determinant of happiness. Since happiness is

often considered to be a plausible proxy for utility it is almost self-evident to address

any presumed link between religiosity and happiness from the perspective of basic

utility theory. However, the theoretical link between happiness and religiosity has not

been established in the literature. The present study aims to fill this gap. We develop

a simple theoretical framework to be used as a point of reference when assessing the

empirical evidence on various links between religiosity and happiness.

The Easterlin paradox has long been held to be at odds with the idea that happiness

is a proxy for utility. Easterlin (1973, 1974, 1995) states that rich people generally

report higher levels of happiness than poor people, but rising average incomes do not

increase happiness beyond a satiation point. Accordingly, an increase in income

beyond the satiation point only seems to shift the reference point within a society,

without affecting utility as proxied by a measure of happiness. This conclusion contra-

dicts textbook utility theory, where changes in income always shift the indifference

curve to a higher level of utility.

This puzzle evoked two theoretical reactions in research on happiness. Frey and

Stutzer (2006) appeal to mistakes in rational decision making in cases where reported

levels of happiness do not correspond with utility maximization. Accepting a better

paid job with higher commuting cost should not result in less happiness, but there

appears to be evidence that it does. Criticizing the assumption that individuals sys-

tematically fail to maximize their utility, Becker and Rayo (2008) argue that measures

* Gundlach: GIGA (German Institute of Global and Area Studies) and University of Hamburg, 20148

Hamburg, Germany. Tel: +49-40-42887438; Fax: +49-40-4107945; E-mail: gundlach@giga-hamburg.de.

Opfinger: Department of Economics, Trier University, 54286 Trier, Germany. For comments and sugges-

tions on earlier versions we are grateful to seminar participants at the ISNE 2010 Conference in Dublin, the

SSE Annual Meeting 2011 in Houston, the Market and Happiness Conference 2011 in Milan, and the

Annual Meeting of the Research Committee on Development Economics 2011 in Berlin, and to Michael

Funke, Olaf Hübler, Bernd Lucke, Martin Paldam, Ulrich Schmidt, Holger Strulik, and an anonymous

referee.

Review of Development Economics, 17(3), 523–539, 2013

DOI:10.1111/rode.12047

© 2013 John Wiley & Sons Ltd

of happiness and the reported Easterlin paradox might not be founded on utility

theory at all. They consider subjective happiness as an argument of the utility function

rather than a direct proxy for utility itself. According to their approach, utility would

remain in the realm of the empirically unknown and a decline of happiness with rising

income could be interpreted as a simple substitution effect.

It is not without irony that the rationalization of the Easterlin paradox by Becker

and Rayo (2008) has been published as a comment to an empirical study by Stevenson

and Wolfers (2008) that rejects the Easterlin paradox. Deaton (2008) and Sacks et al.

(2010) also provide strong evidence for a robust positive link between aggregate indi-

cators of happiness and (log) per capita income across countries and over time. More-

over, the estimated effects of income on happiness closely resemble the well-known

within-country correlation between individual levels of happiness and individual

income.

The new empirical evidence allows for a fresh start of empirical research on happi-

ness and religiosity that is based on a standard model of utility maximization. We

revise the model proposed by Becker and Rayo (2008) by treating happiness as a

direct proxy for utility, which has also been the starting point of the older empirical

happiness literature (Frey and Stutzer 2002). Accordingly, higher income should be

reflected by higher-level indifference curves for happiness. Religiosity enters as one of

the commodities of the happiness function. We show that our theoretical framework

can account for three stylized facts of the empirical literature, namely a positive corre-

lation between happiness and income, a positive correlation between happiness and

religiosity, and a negative correlation between religiosity and income.

2. Basic Results of the Empirical Literature on Happiness and Religiosity

Ellison (1991) divides religious involvement into denominational ties, divine relations,

existential certainty, and social integration. Religious involvement is reported to be

positively correlated with subjective well-being. Similarly, Greene and Yoon (2004)

assert that subjective well-being rises with religious attachment measured by the will-

ingness to attend religious services regularly. Ferriss (2002) confirms a positive corre-

lation between happiness and the frequency of church attendance but points to

denominational and doctrinal differences across churches.

One strand of the literature discusses the effects of religious activities on well-being

over the life cycle. Peacock and Poloma (1999) define religiosity by the four catego-

ries personal devotion, participation in public ritual, divine interaction, and the pref-

erence for public or private religiosity. They suggest that religiosity increases with age

and as such tends to increase reported well-being. Ardelt (2003) uses a survey method

to analyze the relationship between well-being of elderly people and different indica-

tors for intrinsic and extrinsic religious orientation. She shows that religious affiliation

and the frequency of religious attendance reduce the fear and increase the acceptance

of death, but a purpose in life is found to be more important for the well-being of

elderly people than holding religious beliefs per se.1

Lelkes (2006) uses the economic transition in Hungary after the collapse of

socialism as an exogenous shock and corroborates that higher church attendance is

positively correlated with reported well-being. Hayo (2007) also investigates the

determinants of happiness across Eastern Europe after the collapse of the socialist

systems and finds that frequent churchgoers report a significantly higher life satisfac-

tion than those who do not attend church, with no difference in life satisfaction across

different denominations. Elliott and Hayward (2009) use responses from the World

524 Erich Gundlach and Matthias Opfinger

© 2013 John Wiley & Sons Ltd