Voice of the Customers: Local Trust Culture and Consumer Complaints to the CFPB

• Published date: 01 June 2021
• Author: RACHEL M. HAYES,FENG JIANG,YIHUI PAN
• Date: 01 June 2021
• DOI: http://doi.org/10.1111/1475-679X.12364

DOI: 10.1111/1475-679X.12364

Journal of Accounting Research

Vol. 59 No. 3 June 2021

Printed in U.S.A.

Voice of the Customers: Local Trust

Culture and Consumer Complaints

to the CFPB

RACHEL M. HAYES,∗FENG JIANG,†AND YIHUI PAN∗

Received 7 December 2017; accepted 29 January 2021

ABSTRACT

We use complaints filed with the Consumer Financial Protection Bureau

(CFPB) to study the interplay between social norms and the effectiveness of

consumer protection laws. We find that a higher level of trust in a given lo-

cation is associated with a lower number of complaints filed against financial

institutions in that location. Employing a difference-in-differences approach,

we further find that, after the establishment of the CFPB, banks in low-trust

areas reduce fees charged to consumers more compared to banks in high-

trust areas. Our results suggest that the threat of consumer complaints to a

government agency affects how banks treat their customers, and they shed

∗University of Utah; †University at Buffalo

Accepted by Christian Leuz. We would like to thank Rajesh Aggarwal, Zhenhua Chen,

Mike Cooper, Robert DeYoung, Isil Erel, Jun-Koo Kang, Erik Lie, Chen Lin, Karl Lins, Angie

Low, Michelle Lowry,Thomas Schmidt, Sudhir Suchak, Elizabeth Tashjian, KC John Wei, An-

drew Winton, Brian Wolf, and participants of seminars at University of Hong Kong, HKUST,

Hong Kong Polytechnic University, City University of Hong Kong, Nanyang Technology

University, Purdue University, University of Iowa, and University of Utah for their helpful

comments and suggestions. We thank Paola Sapienza and Luigi Zingales for sharing their

Financial Trust Index data, and Francisco Navarrosanchez for excellent research assistance.

An online appendix to this paper can be downloaded at http://research.chicagobooth.edu/

arc/journal-of-accounting-research/online-supplements

Some of the data used in this analysis are derived from Sensitive Data Files of the GSS, obtained

under special contractual arrangements designed to protect the anonymity of respondents.

These data are not available from the authors. Persons interested in obtaining GSS Sensitive

Data Files should contact the GSS at GSS@NORC.org.

1077

© 2021 The Chookaszian Accounting Research Center at the University of Chicago Booth School of

Business

1078 r. m. hayes, f. jiang, and y. pan

light on the interaction between informal culture and formal institutions, as

well as on stakeholders’ influence in corporate policies.

JEL codes: G21, M40

Keywords: culture; trust; Consumer Financial Protection Bureau; bank fees

1. Introduction

In 2010, the Dodd-Frank Act (the Act) authorized the creation of the Con-

sumer Financial Protection Bureau (CFPB), which regulates consumer fi-

nancial products and services, provides consumer education, and most im-

portantly for our purposes in this paper, collects and publishes consumer

complaints to “identify inappropriate practices … [and] stop them before

they become major issues.”1Using data from the CFPB, we first study the

effect of local culture—specifically, the prevailing attitude of trust—on con-

sumer complaints. We then examine whether the threat of consumer com-

plaints to the CFPB affects how banks treat their customers. Answers to

these questions shed light on the interaction between informal culture and

formal institutions and on stakeholders’ influence on corporate policies.

Further, our study contributes to recent political debates about reforms of

the Act.

To guide our analysis, we use a model from Prendergast [2002] featuring

consumer feedback, which has two important testable implications. First,

consumers who are more informed are more likely to complain to the

principal (here, the CFPB). Second, the degree to which consumers are

informed affects the behavior of the agent (the bank). We hypothesize

that the attitude of trust that prevails in a consumer’s local area will affect

how informed a consumer is: Higher trust leads to lower concern for self-

protection, which results in less verification or monitoring by consumers,

and, consequently, less-informed consumers. Our hypothesis is grounded

in the prior literature on trust, which characterizes trust as an expectation

about behavior irrespective of the observability of that behavior. For exam-

ple, Sapienza and Zingales [2012] conduct a survey to measure financial

trust and define trust as “the expectation that another person (or institu-

tion) will perform actions that are beneficial, or at least not detrimental,

to us regardless of our capacity to monitor those actions.” The notion of

trust, as described by Gambetta [1988], is predicated on “the condition

of ignorance or uncertainty about other people’s behavior.” Consumers

with a higher level of trust expect the bank to act in a nondetrimental way

despite their lack of information about the bank’s actions. In contrast, for

consumers with lower trust, the higher perceived likelihood of detrimental

behavior motivates greater concern for self-protection.2Our predictions

1See https://www.consumerfinance.gov/data-research/consumer-complaints/.

2Guiso, Sapienza, and Zingales [2008a] consider the effects of trust on investor self-

protection in the context of equity markets, relating stock market participation to trust, where

local trust culture and consumer complaints to the cfpb

1079

are thus that lower trust levels will be associated with more complaints to

the CFPB and a greater disciplinary effect on bank behavior.

The literature on trust also suggests that trust is partly shaped by a per-

son’s local environment and social networks (Giannetti and Wang [2016];

Gurun, Stoffman, and Yonker [2017]).3Ouranalysisthusincorporatestwo

measures of trust that vary geographically. We begin by measuring trust at

the state level using data from the General Social Survey (GSS), a personal-

interview survey of social characteristics and attitudes. Because the GSS is

conducted biennially, this measure varies both cross-sectionally and across

time. To explore the time-series variation, we consider the revelation of the

Wells Fargo (WF) account scandal in 2013 as a negative shock to trust in fi-

nancial institutions. With the scandal’s revelation, we indeed find a decline

in the level of trust, as measured by the GSS, especially in places with WF

branches. We also use census data to construct a trust measure that takes ad-

vantage of information about the ancestral origins of the local population

(Guiso, Sapienza, and Zingales [2006]). This measure, which is available at

the state, city, and zip code levels, is largely driven by historical immigration

patterns and the attitudes of trust in immigrants’ countries of origin, and

is expected to capture the inherited, persistent component of trust.

In the first part of our study, we test the implication that consumers living

in places where the norm is a lower level of trust file more complaints

to the CFPB. Using both cross-sectional and time-series variation in trust,

we find a robust negative relation between local norms of trust and the

number of consumer complaints. We focus on complaints that were re-

solved with relief—monetary or otherwise—in order to exclude potentially

nonmeritorious complaints.4This negative relation holds across states, as

well as across a bank’s branches in different cities or zip codes within a

state. It also holds when we compare (non-WF) complaints before and

after the revelation of the WF account scandal across locations that are

affected by the revelation to different degrees. In additional analysis, we

attempt to separate the effect of local consumer trust from the potential

confounding effects of bank employee culture in the same location. We

examine complaints related to credit cards, which are usually marketed to

consumers nationally, whereas the influence of the card business’ local cul-

ture on bank employees’ behavior will be captured by bank fixed effects.5

trust is defined as the perceived likelihood of deception or expropriation. Christensen, Maf-

fett, and Vollon [2019} further study how security regulations interact with trust to affect

country-level household equity ownership.

3For example, see Lins et al. [2017} on the role of trust between firms and their stakeholders

and investors.

4Complaints that do not receive monetary or other relief may result, at least in part, from

consumers’ lack of financial literacy. Because financial literacy may vary cross-sectionally, we

conduct our tests using complaints that are less subject to this concern. In section 3.2, we

use enforcement action data to provide further evidence for the merit of the complaints

we consider.

5Credit card operations are highly clustered in Delaware and South Dakota.