The Influence of Affect on Heuristic Thinking in Insurance Demand

• Date: 01 March 2017
• Author: Johannes G. Jaspersen,Vijay Aseervatham
• DOI: http://doi.org/10.1111/jori.12088
• Published date: 01 March 2017

©2015 The Journal of Risk and Insurance. Vol.84, No. 1, 239–266 (2017).

DOI: 10.1111/jori.12088

The Influence of Affect on Heuristic Thinking in

Insurance Demand

Johannes G. Jaspersen

Vijay Aseervatham

Abstract

Heuristic thinking can influence human behavior in decisions under risk and

uncertainty.In an experimental setting, we study whether emotional activa-

tion primes individuals to use the representativeness heuristic and the affect

heuristic. We observe the decision behavior of 272 subjects in a computer-

based experiment that differentiates between incidental affect and integral

affect. Positive incidental affect and integral affect increasethe use of the rep-

resentativeness heuristic, while negative incidental affect has no effect. Our

findings have statistical and economic significance and carry implications

for insurance companies and regulators.

Introduction

A variety of studies have shown that people use heuristics when making decisions

under risk (Clotfelter and Cook, 1993; Kahneman, Slovic, and Tversky, 1982; Lacetera,

Pope, and Sydnor, 2012). However,while many studies have been conducted in order

to identify heuristic patterns, up until now, researchers have mostly refrained from

trying to determine the cause for the utilization of heuristics (for a notable exception

see, e.g., Payne, Bettman, and Johnson, 1993). In insurance demand, similarly, there

have been documentations of heuristic thinking (e.g., Brown, 2007; Ganderton et al.,

2000; Papon, 2008), but no analyses of what causes heuristic processing.

Johannes G. Jaspersen and Vijay Aseervatham areat the Ludwig-Maximilians-Universität Mu-

nich, Munich School of Management, Munich, Germany. Jaspersen and Aseervatham can be

contacted via e-mail: jaspersen@bwl.lmu.de and aseervatham@bwl.lmu.de, respectively.Finan-

cial support from LMUexcellent is gratefully acknowledged. We kindly thank the Munich Ex-

perimental Laboratory for Economic and Social Sciences for providing laboratory resources.We

want to thank the editor of the Journal of Risk and Insurance, Keith Crocker,and three anonymous

reviewers for providing very helpful comments. We are also thankful for the comments from

participants at the 2011Hamburg-M ¨

unchen-Hohenheim Insurance Economics Colloquium, the

2011 MELESSA Brownbag Seminar, the 2012 Annual Meeting of the German Insurance Science

Association, the 2012 Annual Meeting of the American Risk and Insurance Association, the

2012 Annual Meeting of the German Finance Association, the 2012 CEAR/MRIC Behavioral

Insurance Workshop, and the 2013 Annual Meeting of the Asia-Pacific Risk and Insurance

Association. In particular, we are indebted to Jennifer Coats, Pascal F¨

orster, Theresa Michl,

Charles Noussair, Andreas Richter, Harris Schlesinger, Justin Sydnor, Sharon Tennyson, and

Peter Wakker for valuable input. All remaining errorsare ours.

239

240 The Journal of Risk and Insurance

Neuroeconomic research implies that cognitive inferences like heuristics can often

be accompanied by underlying emotional, also called affective, processes (Damasio,

1994; De Martino et al., 2006; Trepel, Fox, and Poldrack, 2005). Hogarth et al. (2011),

for example, find that emotional states explain significant variance in everyday risk

perception. Regarding the decision process, increased emotional arousal has been

shown to reduce the cognitive capacity of an individual (Pham, 2007), which favors

heuristic thinking over more substantive processing strategies (Forgas, 1995). Thus,

exploring the emotional component of decision situations might shed some light on

when and for which reasons heuristics are used in insurance decisions.

In this article, we offer first evidence that emotions can influence whether people

use heuristics while purchasing insurance. For this, we conduct a computer-based

laboratory experiment. Our subjects make 20 consecutive insurance choices with equal

incentives in each decision problem. We investigate the behavior of individuals after

a loss event to observe the representativeness heuristic in the form of the gambler’s

fallacy (Clotfelter and Cook, 1993). We also manipulate subjects with regard to their

emotions toward the item to insure. This lets us observe the utilization of the affect

heuristic (Finucane et al., 2000). We induce experimental subjects with joy or sadness

to stimulate, respectively mute, their emotional activation. These treatments let us

determine the influence of affective activation on the utilization of both heuristics. We

can also observe whether the emotional activation of the item to insure leads to an

increased utilization of the gambler’s fallacy.

Our results show that whether or not people are emotionally activated influences the

utilization of the gambler’s fallacy when determining insurance demand after a loss

event. Specifically, subjects who have been induced with joy or an integral affective

response toward the item to insure use the gambler’s fallacy more than nonactivated

subjects. Furthermore, we argue that the affect heuristic works differently than the

cognitive gambler’s fallacy.

Understanding demand patterns for insurance is very important for both insurance

companies and the regulator. Insurance companies spend considerable resources on

marketing in order to influence consumer behavior. Much of this money is spent on

untargeted messages such as television spots or print advertisements. A better under-

standing of when consumers use heuristic demand patterns and when they use more

substantial cognitive processing strategies might help insurance companies to spend

their money in a more targeted and efficient way and might thus increase revenues

and profitability. The regulator could be interested in what causes heuristic thinking

in insurance demand because they might want to inhibit specific marketing strategies

of insurance companies. This would particularly be the case if those strategies are

exploiting consumers.

The rest of the article is structured as follows. The second section offerscoverage of the

theoretical framework underlying our analysis: the affect infusion model (AIM). We

also give a literature review of studies regarding the gambler’s fallacy and the affect

heuristic in insurance demand decisions. The third section describes the experimental

setup for our study. We develop hypotheses in the fourth section. The fifth section

provides an overview of the results. These and their implications are discussed in the