External Rotation and Auditor-Client Negotiations: The Role of Affective Relationship Commitment and Power

How to cite

Elten, Jonas van; Koch, Christopher. 2015. "External Rotation and Auditor-Client Negotiations: The Role of Affective Relationship Commitment and Power." AEA RCT Registry. AEARCTR-0000897. https://doi.org/10.1257/rct.897-2.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Jonas van Elten* , Christopher Koch

* Primary investigator

Registry ID
AEARCTR-0000897
DOI
https://doi.org/10.1257/rct.897-2.0
First registered
Last updated
Status
In development
Trial dates
2015-10-06 – 2015-10-30
Countries
Private
JEL codes
C78, C91, C92
Registry record
http://www.socialscienceregistry.org/trials/897

Abstract

The European Union has recently introduced an external rotation requirement to enhance auditor independence and audit quality. Prior literature argues that audit firm rotation may enhance auditor independence because it reduces economic incentives to retain the client. This study goes beyond the economic effects and considers the effects of the rotation requirement on the personal relationship between the auditor and the client. Based on the dual concern model, we argue that a weaker personal relationship between both parties due to the rotation requirement can hinder effective auditor-client negotiations. In contrast, the personal relationship that develops when the auditor is retained for several periods can enable integrative solutions. Moreover, we argue that the positive effects of a personal relationship can be realized especially when the auditor has sufficient power. To test our hypotheses, we follow an experimental economics approach. This approach enables us to disentangle financial incentives from affective commitment arising from a personal relationship.

Intervention

The experiment is a 2 by 2 factorial design with the independent variables rotation regime and auditor power fully crossed.

We manipulate the rotation regime through the number of periods auditors interact with clients. Investors and managers remain in the same group throughout all 15 experimental rounds. In the rotation condition, the auditor is randomly assigned to a new investor-manager pair after each round. In the retention condition, the auditor interacts with the same investor-manager pair throughout the whole experiment.

We manipulate auditor power by varying the economic disadvantages managers experience from a qualified audit report. In the high-power condition, auditor rejections reduce manager payoffs by 50%. In the low-power condition, auditor rejections reduce manager payoffs by 5%.

Experimental design

We implement an experimental economics approach with incentivized multi-period interaction among student-subjects. The experiment is designed as a modified version of the standard two-player trust game. In a trust game, investor-subjects receive an initial endowment and decide how much of their endowment to send to an anonymous responder-subject. The responder receives a multiple share of the investment and likewise decides which share to return. In the modified version, the decision of the responder is subject to approval by an independent supervising body that we refer to as gatekeeper. The gatekeeper reflects an external auditor who negotiates with the responder about the return on investment and decides on behalf of the investor whether the returned share is appropriate or not.

We capture the modified trust game in a five-stage experimental design. Before the experiment begins, student-subjects are randomly assigned to one of the three experimental roles as investor, manager, or auditor. In the first stage, investor-subjects receive an initial endowment and decide whether or not to entrust resources to the manager. Managers receive a multiple share of the investment in the second stage, and communicate their preferred distribution to the auditor. In the third stage, auditors may convince the manager to return higher shares to the investor by sending the manager text-messages. We refer to the interaction between auditors and managers as negotiation process. In the forth stage, managers receive the auditor's message and decide which share to return to the investor. Finally, in the fifth stage, auditors accept or reject the manager's return on behalf of the investors. This five-stage trust game describes one experimental round. Each subject participates in 15 experimental rounds.

Hypotheses:

H1a: Auditors apply integrative negotiation strategies less frequently when audit-firm rotation is mandated than when its not.
H1b: Auditors reject outcomes that fall short of their negotiation aim more often when audit-firm rotation is mandated than when its not.

H2a: Auditors apply contending negotiation strategies more frequently when their power position is strong than when its weak.
H2b: Auditors reject outcomes that fall short of their negotiation aim more often when their power position is strong than when its weak.

H3a: Auditors apply integrative negotiation strategies most often when their power position is strong and when audit-firm rotation is not mandated.
H3b: Auditors reject outcomes that fall short of their negotiation aim most often when their power position is strong and when audit-firm rotation is not mandated.

We conduct the experiment in the Experimental Economics Laboratory at the University of Mannheim (mLab) and the Frankfurt Laboratory for Experimental Economics at the Goethe University of Frankfurt (FLEX). The experiment is programmed and conducted with the software z-Tree and subjects are recruited from the general student population using ORSEE.

Design details

Randomization method
Randomization is done by the experimental software z-Tree.
Randomization unit
Student-subjects are randomly seated, randomly assigned to their experimental role, and randomly divided into groups of three players.
Clusters
A total of 240 student-subjects participate in the experiment.
Observations
80 auditor-subjects participate in 15 experimental rounds each.
Arms
The experiment is a fully crossed 2*2 factorial design. 60 student-subjects participate in each of the four experimental conditions.
Primary outcomes
The main dependent measure is the auditor's negotiation strategy. It is specified by the number of messages auditors choose from the five available strategy options presented in a negotiation catalogue: avoiding, conceding, contending, compromising, or integrative strategies.

Cited by

No citation data available for this registration yet.

Keywords