Shrouding Add‐On Information: An Experimental Study*

• Published date: 01 October 2019
• Date: 01 October 2019
• DOI: http://doi.org/10.1111/sjoe.12319

Scand. J. of Economics 121(4), 1705–1727, 2019

DOI: 10.1111/sjoe.12319

Shrouding Add-On Information: An

Experimental Study*

Hans-Theo Normann

D¨usseldorf Institute for Competition Economics (DICE), DE-40225 D¨usseldorf, Germany

normann@dice.hhu.de

TobiasWenzel†

University of Sheffield,Sheffield, S1 4DT, UK

t.wenzel@sheffield.ac.uk

Abstract

We explore how increased competition affects firms’ obfuscation strategies in a laboratory

experiment. Firms sell a base good and an add-on product. Besides choosing the base-good

price, sellers take an action that mimics the effects of shrouding the add-on product. Shrouding

is an equilibrium but an unshrouding equilibrium coexists. In our experiment, more competition

matters, in that only duopolistic markets are frequently shrouded whereas four-firm markets

are not. With repeated interaction, shrouding rates do not increase. However, the opportunities

to shroud facilitate tacit collusion on the base-good price for the duopolies: the unshrouding

equilibrium serves as a credible punishment if deviations occur.

Keywords: Add-on price; non-attentive consumers; shrouding

JEL classification:C7; C9; L4; L41

I. Introduction

For behavioral industrial organization, a key question is to what extent

firms can exploit consumer irrationalities. It is, by now, well accepted that

consumer decision-making is far from perfect. Consumers make mistakes,

can use simple (non-optimal) rules of thumb (Ellison, 2006), and are subject

to behavioral biases (DellaVigna, 2009). How rational firms respond to

those consumer behaviors is at the heart of recent behaviorally founded

theories in industrial economics (see Ellison, 2006; Spiegler, 2011).

Obfuscation strategies are a prime example of this kind of behavioral

industrial organization, which firms can use to target myopic, inattentive

*We would like to thank Steve Davies, Yiquan Gu, and Chris Wilson, as well as seminar and

conference participants at Aachen, Duisburg,D ¨usseldorf, Liverpool, Loughborough, and WHU,

and participants at RES (Manchester), IIOC (Washington), and ESA (Santa Cruz), for very

helpful comments and suggestions. Weare also grateful to three anonymous referees for valuable

suggestions that have substantially improvedthe paper.

†This work was completed whileT. Wenzelwas at the University of Bath.

C

The editors of The Scandinavian Journal of Economics 2018.

1706 Shrouding add-on information

consumers. Consumer myopia has been observed, for example, in financial

markets (e.g., Campbell, 2006; Choi et al., 2010), electricity markets

(Wilson and Waddams Price, 2010), and online auctions (Hossain and

Morgan, 2006). In these examples, consumers do not choose optimally

but myopically pick a suboptimal tariff or inefficient bid. The examples

also show that firms can respond with some obfuscation strategy. They

can highlight irrelevant information (Choi et al., 2010), develop redundant

financial innovations (Henderson and Pearson, 2011), or shroud certain price

elements (Campbell, 2006).

It is more difficult to answer the question of whether firms’ attempts

to exploit myopic consumers are successful, not least because competition

is a forceful argument suggesting this might not be the case. However, in

a pioneering and frequently cited paper, Gabaix and Laibson (2006) show

that intensified competition might not have any effect. In a model with

rational (attentive) and myopic (inattentive) consumers, they show that the

shrouding of add-on information can be an equilibrium despite perfect price

competition.

In this paper, we conduct a laboratory experiment to explore the effects

of increased competition on shrouding behavior. We consider a stylized

version of Gabaix and Laibson (2006) where firms sell a base good and

an add-on product. Firms decide on the base-good price as well as on an

action that mimics the effects of shrouding the add-on. This decision is a

coordination decision where the add-on is shrouded to consumers only if

all firms decide to shroud. Otherwise, if at least one firm unshrouds, all

consumers are perfectly informed about the add-on. We consider both a

static game where firms only interact once as well a dynamic game where

firms repeatedly compete over a finite number of periods. Both shrouding

and unshrouding equilibria coexist in the static game.

An experimental approach seems well suited for exploring this issue

because theory is often bland regarding equilibrium multiplicity (the

shrouding equilibrium exists for any number of firms). We implement

a stylized design with a focus on the firm side and where the buyer

side is computerized (see our literature survey below for experiments

analyzing the buyer side). We consider the following treatments. The degree

of competition is varied by changing the number of firms: we conduct

experiments with two and four sellers. A second treatment variable is

the matching scheme: we have markets with random and fixed matching

to test the implications of the static game as well as of the repeated

game. Repeated interaction itself can lead to improved coordination in a

shrouding equilibrium. Moreover, one of our main goals is to study how

the shrouding and the pricing decision affect one another. Therefore, we

additionally examine two treatments that study either the shrouding or the

C

The editors of The Scandinavian Journal of Economics 2018.