DOES RELIGIOSITY AFFECT SUPPORT FOR POLITICAL COMPROMISE?

• Date: 01 August 2016
• Published date: 01 August 2016
• Author: Danny Cohen‐Zada,Oren Rigbi,Yotam Margalit
• DOI: http://doi.org/10.1111/iere.12186

INTERNATIONAL ECONOMIC REVIEW

Vol. 57, No. 3, August 2016

DOES RELIGIOSITY AFFECT SUPPORT FOR POLITICAL COMPROMISE?∗

BYDANNY COHEN-ZADA,YOTAM MARGALIT,AND OREN RIGBI1

Ben-Gurion University, Israel; Tel-Aviv University, Israel; Ben-Gurion University, Israel

Does religiosity affect adherents’ attitude toward political compromise? To address this question and overcome the

potential simultaneity of religious activity and political attitudes, we exploit exogenous variation in the start date of the

Selichot (“Forgiveness”), a period in which many Jews, including nonadherents, take part in an intense prayer schedule.

Using a two-wave survey, we find that an increase in the salience of religiosity leads to the adoption of more hard-line

positions against a land-for-peace compromise. Examining several potential mechanisms for this attitudinal shift, our

evidence points to the impact of the intensified prayer period on adherents’ tolerance for risk.

1. INTRODUCTION

Does religious activity affect individuals’ attitudes toward political compromise? The impact

of religiosity on political attitudes is a long-standing and contested issue. Whereas a long line of

thinkers have emphasized the pacifying role of religious fervor in promoting such traits as regard

for others and compassion (Freud, 1927; Skinner, 1969), others contend that religiosity often

breeds intolerance, bigotry, and intergroup hostility (e.g., Allport, 1954; Stouffer, 1955; Dawkins,

2003). Concerns about the impact of religiosity on decreasing willingness to accept compromise

have led analysts to warn about the negative consequences of long-standing national–territorial

conflicts—such as those over Kashmir or Palestine—from transforming into religious ones. The

claim is that the infusion of religion into territorial disputes transforms them into ones fought

over nonnegotiable absolutes, making political compromise much harder to attain (Hassner,

2009). Indeed, prior research indicates that conflicts in which warring factions couched their

claims in explicit religious terms are significantly less likely to be terminated through negotiated

settlement (Svensson, 2007).

Investigating the barriers to peaceful resolution of such political conflicts, including the

possible role of religiosity, is warranted not only because of the terrible cost in lives that

prolonged conflicts cause, but also the substantial economic price that the warring populations

incur. Collier (1999) estimates, based on cross-national data, that each conflict year accounts

for an average 2.2 percentage point loss in GDP for a country engaged in intra-state war.2

Case studies of specific conflicts further demonstrate the sizable economic losses resulting from

armed clashes. For example, Abadie and Gardeazabal (2003) estimate that terrorist attacks in

the Basque region of Spain in the late 1960s led to a loss of about 10% of per capita GDP.

Eckstein and Tsiddon (2004) conclude that the wave of attacks in the second Palestinian uprising

(“Intifada”) between 2001 and 2003 led to an annual decline of over 3% in output per capita. A

∗Manuscript received October 2013; revised February 2015.

1For helpful comments, we thank Leif Danziger, Liran Einav, Mridu Prabal Goswami, Shigeo Hirano, and David

Voas. Rigbi gratefully acknowledges financial support from the European Community’s Seventh Programme (grant

no. 249232) and the Israeli Science Foundation (grant No. 1255/10). Cohen-Zada and Rigbi gratefully acknowledge

financial support from the Maurice Falk Institute for Economic Research in Israel. Please address correspondence to:

Oren Rigbi, Department of Economics, Ben-Gurion University of the Negev, P.O.B 653, Beer-Sheva 84105, Israel.

E-mail: origbi@bgu.ac.il.

2Estimates of conflict costs can differ substantially, even within a single case. In part this is a function of the way

in which costs are measured (by physical destruction, reduced private investment, loss of human capital, or shifts in

budget allocations), but it is also due to a difference in method (e.g., accounting, counterfactual analysis).

1085

C

(2016) by the Economics Department of the University of Pennsylvania and the Osaka University Institute of Social

and Economic Research Association

1086 COHEN-ZADA,MARGALIT,AND RIGBI

recent report by the Israeli Finance Ministry strengthens the findings: It estimates that attaining

peace with the Palestinians would provide the Israeli government with savings amounting to

more than 7.5% of its annual spending and boost annual exports by almost 5%.3

Yet the projected economic dividends from ending conflict do not necessarily translate into

public support for political compromise. In Israel, the country studied in this article, a land-for-

peace compromise with the Palestinians is supported by roughly half of the voting public, with

opposition particularly high among religious voters (The Peace Index, 2014).4Indeed, in recent

decades religious voters have come to represent a central component of the peace-skeptic right

wing bloc (Shamir and Arian, 1999; Shelef and Shelef, 2013). But to what extent does religiosity

itself underlie their opposition to political compromise?

The answer is far from obvious. In fact, earlier efforts to empirically investigate the relation-

ship between religiosity and political attitudes have produced decidedly ambiguous findings.

Because religiosity is not randomly assigned across the population, a correlation between re-

ligious beliefs and certain political attitudes does not necessarily imply causality, as the two

may be co-determined by other underlying factors.5Although many previous studies have

recognized the endogeneity of religiosity, only a few have attempted to account for it.6

In this study, we examine the impact of heightened religiosity on the attitudes of Israeli Jews

toward political compromise with the Palestinians. We focus on changes in attitudes that arise as

a result of religious activity during the days of Selichot (“forgiveness”), a period of intense prayer

that takes place during the final month of the Jewish calendar year. To deal with the empirical

challenge of identifying the influence of religious activity on political attitudes, we exploit two

unique features of the Selichot period. First, Judaism’s two main ethnic traditions dictate a

slightly different start date for the prayer schedule. As a result, during the first weeks of the

Selichot, people of the same religion, who live side by side and who otherwise share similar levels

of religious observance, suddenly experience very different levels of prayer activity for reasons

unrelated to the strength of their religious belief. Second, it is customary during the Selichot for

even irregular observers to attend an intense schedule of prayer.

To examine whether heightened religiosity affects adherents’ views on a land-for-peace com-

promise, we compare the views among members of the different ethnic groups across two points

in time: before the Selichot begins for both groups and after the Selichot begins for just one of

the groups. This design allows us to compare the difference in attitudes both within and across

groups over time.

We find that the Selichot leads to an average increase of about 17 percentage points in the

probability of strongly opposing a land-for-peace compromise. We also find that the Selichot is

associated with a 18-percentage-point drop in the probability of an individual expressing strong

support for territorial compromise. In substantive terms, the effects are approximately 30%

and 45% as large as the difference in views between left- and right-wing voters, respectively.

These effects also hold when controlling for a host of individual-level characteristics. Although

our study is not a panel design, the main pattern of change across the two waves appears to

reflect a rightward shift in attitudes within each ideological block: Instead of switching from

support for territorial compromise to opposition, the evidence shows a growing skepticism of

compromise both among the camp originally in favor of an agreement (i.e., a shift from strong

to weak support) as well as among the camp opposed to an agreement (i.e., a shift from weak

to strong opposition).

3See, “Lapid: If talks with Palestinians collapse, economy will be battered,” Haaretz, 03/01/2014. In absolute terms,

this amounts to $5.7 billion in annual spending and an increase of almost $4.6 billion in exports.

4According to the September 2014 Peace Index publication, a monthly polling report that tracks Israeli public

opinion, 65.4% of self-identified religious respondents expressed opposition to negotiating a peace agreement with the

Palestinian Authority, whereas only 24.8% of self-identified secular respondents opposed such an agreement.

5Indeed, Cohen-Zada and Sander (2011) show that religious participation is correlated with a broad range of observed

individual characteristics, indicating that unobserved characteristics may be a major problem for causal identification.

6See Gruber, 2005; Gruber and Hungerman, 2008; Clingingsmith et al., 2009; Cohen-Zada and Sander, 2011; Lee,

2013; and Gerber et al., 2016.