The Market for Conflicted Advice

• Author: MARTIN SZYDLOWSKI,BRIANA CHANG
• Published date: 01 April 2020
• Date: 01 April 2020
• DOI: http://doi.org/10.1111/jofi.12848

THE JOURNAL OF FINANCE •VOL. LXXV, NO. 2 •APRIL 2020

The Market for Conflicted Advice

BRIANA CHANG and MARTIN SZYDLOWSKI∗

ABSTRACT

Wepresent a model of the market for advice in which advisers have conflicts of interest

and compete for heterogeneous customers through information provision. The com-

petitive equilibrium features information dispersion and partial disclosure. Although

conflicted fees lead to distorted information, they are irrelevant for customers’ welfare:

banning conflicted fees improves only the information quality,not customers’ welfare.

Instead, financial literacy education for the least informed customers can improve all

customers’ welfare because of a spillover effect. Furthermore, customers who trade

through advisers realize lower average returns, which rationalizes empirical findings.

PROVIDING ADVICE IS AN IMPORTANT function of intermediaries. The compensa-

tion structure of intermediaries, however, often leads to conflicts of interest.

For example, broker-dealer firms are compensated by commissions and fund

distribution fees, and realtors receive fees only if they close a deal. Financial ad-

vice has received considerable attention from academics and regulators alike.

It is well documented that conflicted advice leads to lower investment returns,

which imposes a substantial loss on households.1As a result, many countries

have started regulating adviser fees in an effort to eliminate conflicted advice.

∗Briana Chang is with the University of Wisconsin-Madison. Martin Szydlowski is with the Uni-

versity of Minnesota. We thank the Editor,Philip Bond, as well as an anonymous associate editor

and two anonymous referees for helpful comments. Wewould also like to thank Dean Corbae; Mark

Egan; Simon Gervais; Erica Johnson; Pablo Kurlat; Ricardo Lagos; Gregor Matvos; John Nash;

Dimitry Orlov; Christine Parlour; Erwan Quintin; Adriano Rampini; Mark Ready; S¨

onje Reiche;

G¨

unther Strobl; Vish Viswanathan; and Randy Wrightfor their useful discussions; as well as par-

ticipants at the 8th Annual Conference on Money, Banking and Asset Markets, the Asian Meeting

of the Econometric Society,the Financial Intermediation Research Society Conference, the Finance

Theory Group Meeting, the Frankfurt School of Finance and Management, the Midwest Economic

Theory Conference, the 9th NYU Search Theory Workshop, the Society for the Advancement of

Economic Theory Conference, the Society for Economic Dynamics Annual Meeting, the Society for

Financial Studies Cavalcade, the ToulouseSchool of Economics, the UNC/Duke Corporate Finance

Conference, the University of Minnesota, the University of Wisconsin, the West Coast Search and

Matching Workshop, and the WU Gutmann Center Symposium for helpful comments. The authors

do not have any conflicts of interest, as identified in The Journal of Finance’s disclosure policy.

1Empirical literature shows that financial advisers lead customers to chase returns (Linnain-

maa, Melzer,and Previtero (2018)), steer customers toward high-fee, actively managed investments

(Mullainathan, Noeth, and Schoar (2012)), and recommend unsuitable products when advisers

earn high commissions (Anagol, Cole, and Sarkar (2017)), and that in consequence, portfolios of

advised customers underperform (e.g., Bergstresser, Chalmers, and Tufano(2009), Chalmers and

Reuter (2010), Hoechle et al. (2017)). The Council of Economic Advisers (CEA (2015)) further find

a substantial welfare loss from conflicted advice in the United States. As a result, the Department

DOI: 10.1111/jofi.12848

C2019 the American Finance Association

867

868 The Journal of Finance R

However, how markets for advice function and their consequences for investors

remain unknown. To design effective policy, it is necessary to understand why

customers find it optimal to trade through advisers and how advisers respond

to regulation.

In this paper, we develop an equilibrium model in which the quality of advice,

customers’ investment decisions, and their choice of advisers are determined

jointly in a frictionless matching market. We then use this model to analyze

how regulation may affect customers’ welfare, taking into account how advisers

adjust their quality of advice in equilibrium. To capture advisers’ conflict of

interest, we assume that advisers (which we take to be brokers, retail financial

advisers, or realtors) are compensated only if they successfully convince their

customers to invest. This setup reflects the fact that brokers are compensated

by distribution fees, which are set by fund issuers, and do not charge a separate

fee for providing advice.2,3In our setting, advisers compete for customers by

providing information. Taking these misaligned incentives as given, we aim to

understand the extent to which advisers can generate valuable information for

customers in this competitive setting.

To shed further light on how such conflicts of interest affect different agents

in the economy, we account for both customer and adviser heterogeneity. Cus-

tomers differ in the quality of their ex ante information, while advisers differ in

the value of their expertise. More precisely, each adviser is perfectly informed

about one particular type of asset and assets differ in their information sen-

sitivity. Thus, an adviser with expertise in a more information-sensitive asset

has, in effect, more valuable information.

In our setting, customers understand the misaligned incentives of advisers

and optimally select their advisers based on the information provided in equi-

librium. After choosing an adviser, the customer receives information, ratio-

nally updates his beliefs, and makes his investment decision. Since providing

advice takes time, each adviser is subject to a capacity constraint. For sim-

plicity, we assume that each adviser can match with at most one customer

and vice versa. Advisers are scarce, so not all customers can receive advice.

of Labor has instituted a new rule that holds financial advisers to the fiduciary duty standard. Mo-

tivated by similar concerns, regulators in other countries have enacted measures such as banning

payments from product providers to advisers (Australia, the Netherlands, the United Kingdom) or

mandating disclosures about conflicts of interest (Canada, Germany).

2See, for example, Understanding Your Brokerage and Investment Advisory Relationships by

Morgan Stanley (December 2014): “In addition to taking your orders, executing your trades and

providing custody services, we also provide investor education, investment research, financial

tools and professional, personalized information about financial products and services, including

recommendations to our brokerage clients about whether to buy, sell or hold securities. We do not

charge a separate fee for these services because these services are part of, or ‘incidental to,’ our

brokerage services.” Similar pay structures are common for other types of financial advisers and

may take the form of sales targets, commissions, or kickbacks. See, for example, table 3 in CEA

(2015) for a detailed overview.

3However, another type of financial advisers charge flat fees for their services. Wecompare our

results to these fee-only advisers in Section III.A.

The Market for Conflicted Advice 869

Thus, our setting can be understood as a matching model with two-sided het-

erogeneity, where information provision determines the gains for both sides.

The key equilibrium objects are the distribution of information quality, the

value of information to customers, the profits of advisers, and the matching

patterns.

We derive a notion of information quality that arises endogenously from

the adviser’s optimal choice of what information to provide. We show that

information is dispersed and distorted because of the conflicted fee structure.

Intuitively, since brokers are compensated by fee-upon-investment, they have

an incentive to provide biased advice to extract higher profits. However, to

attract a particular customer, an adviser must provide sufficiently valuable

information that the customer does not prefer to match with someone else. This

competition disciplines advisers and forces them to provide at least partially

informative advice. Specifically, the least informed customers are the easiest

to deceive, and all advisers compete for those customers. Thus, the gain from

trading through advisers, relative to trading on their own, is higher for the

less informed customers. Nevertheless, they still receive worse information in

equilibrium and earn lower investment returns, compared to more informed

customers. Our results therefore suggest that investment returns per se do not

represent the value of advice.

We use our model to establish important results about regulation. Many

countries (e.g., Australia, Italy, the United Kingdom, and the Netherlands; see

CEA (2015, p. 25)) have started to restrict adviser fees based on the idea that

if fees are driving the conflict of interest, capping or eliminating them should

improve the advice provided to customers. However, this argument fails to ac-

count for how advisers might change their information policy in equilibrium.

We show that when regulators lower advisers’ fees, advisers provide worse in-

formation. This is because the utility of a customer in a matching equilibrium

is uniquely pinned down such that his matched adviser would not profit from

attracting his next-best competitor. Since fees and information quality are sub-

stitutes from customers’ perspective, whether customers are compensated by

better information or lower fees is irrelevant. We thus establish that when ad-

visers are on the short side of the market, customers’ welfare is independent of

the fee structure—that is, surprisingly, regulating adviser fees do not improve

customers’ welfare.4

So what can regulators do? Since the utility of a customer is pinned down,

what matters is the underlying distribution of financial literacy, that is, the

distribution of customers’ informedness. Financial literacy education has a

positive spillover effect. Specifically, when some customers in the economy

become better informed, other customers also benefit, even though the lat-

ter’s level of informedness does not change. Intuitively, when a customer

4We consider the case in which advisers are on the long side in the Internet Appendix. In this

case, the additional gain of a customer relative to his next-best competitor is still independent

of the fee structure. Nevertheless, a higher fee decreases all customers’ welfare by the same

amount. The Internet Appendix is available in the online version of this article on The Journal of

Finance website.