A multi-method approach to identifying norms and normative expectations within a corporate hierarchy: Evidence from the financial services industry

This paper presents the results of a field study at a large financial services firm that combines multiple methods, including two economic experiments, to measure ethical norms and their behavioral correlates. Standard survey questions eliciting ethical evaluations of actions in on-the-job ethical dilemmas are transformed into a series of incentivized coordination games in the first experiment. We use the results of this experiment to identify the actual ethical norms for financial adviser behavior held by key personnel - financial advisers and their corporate leaders - in three settings: a clash of incentives between serving the client and earning commissions, a dilemma about fiduciary responsibility to a client, and a dilemma about whistle-blowing on a peer. We also measure the beliefs of financial advisers about the ethical expectations of their corporate leaders and the beliefs of corporate leaders about financial adviser norms. In addition, we ask financial advisers about their personal normative opinions, matching a common methodology in the literature. We find, first, systematic agreements in the normative evaluations across the corporate hierarchy that are consistent with ex ante expectations, but second, we also find some measurable differences between the normative expectations of corporate leaders about on-the-job behavior and the actual norms shared among financial advisers. When there is a normative mismatch across the hierarchy we are able to distinguish miscommunication from ethical disagreement between leaders and employees. Our subjects also report their job satisfaction and take part in a second incentivized experiment in which it is costly to report private information honestly. A last finding is that a mismatch between advisers' personal ethical opinions and corporate norms - especially those of peers - strongly correlates with job dissatisfaction, and less strongly but significantly with the willingness to be dishonest.