Strategic complexity in investment management
fee disclosures

Leslie A. Muller, Ph.D.a,*, John A. Turner, Ph.D.b

aAssistant Professor of Economics, Grand Valley State University, 50 Front Street SW, Grand Rapids, MI
49504, USA

bPension Policy Center, 3713 Chesapeake St. NW, Washington, DC 20016, USA

Abstract

This article develops a measure of complexity of fee disclosures, based on previous work assessing
the grade level of language, and validates that measure through a survey where students are asked to
independently rate the complexity of fee disclosures. In addition, the article hypothesizes that high fee
providers are more likely to engage in strategic complexity in fee disclosures than are low fee
providers. We hypothesize that high fee providers use strategic complexity to take advantage of the
lack of financial sophistication of many people, making it difficult for them to compare fees across
service providers and to understand the level of fees they are paying. © 2016 Academy of Financial
Services. All rights reserved.

JEL classification: G02; G2

Keywords: Fees; Complexity; Financial literacy; Disclosures

1. Introduction

While much attention has been given to financial education for pension participants
and their lack of financial sophistication (e.g., Lusardi and Mitchell, 2007; McCarthy and
Turner, 2000), less attention has been given to the quality of information they receive
from financial service providers. Efforts by pension participants and other investors to
learn about financial market products and services may be offset by efforts by financial

* Corresponding author. Tel.: �1-616-331-7473; fax: �1-616-331-7421.
E-mail address: mullerle@gvsu.edu (L. A. Muller)

Financial Services Review 25 (2016) 215–234

1057-0810/16/$ – see front matter © 2016 Academy of Financial Services. All rights reserved.



service providers to obfuscate through complexity in fee structures and in financial
disclosures concerning fees.

While transparency is widely viewed as desirable in financial disclosures and in the
features of financial products, to take advantage of the lack of financial sophistication of their
clients, some advisers may engage in strategic complexity. With strategic complexity, they
structure their fees and fee disclosures in complex ways. For example, they may disclose
important information in footnotes or use terminology that is not commonly understood by
pension participants. Abeler and Jäger (2015) comment that there is growing evidence that
people often do not respond optimally when they are faced with complex financial
information.

Complexity is a multidimensional strategy that we hypothesize is used by some invest-
ment management or advisory companies to make it difficult to compare advisory services
on the basis of fees, and to determine what level of fees a prospective investor would be
paying. The strategy takes advantage of the lack of financial sophistication of many inves-
tors, but in some instances even sophisticated investors may not be able to make comparisons
or obtain accurate fee information based on the disclosures provided on the Internet.
“Financial sophistication” can come about through taking specific courses (e.g., finance class
in college), life experience in making financial choices, and/or working in a position that
requires financial decision-making. Strategic complexity presumably reduces competition
based on fees in the market for advisory services and permits service providers to charge
higher fees. Prices and fees are essential information in the functioning of the market for a
product or service. Strategic complexity in the disclosure of fees is presumably designed to
make a market less competitive to increase the income of service providers.

Strategic complexity raises search costs. It may explain in part the limited search that
many people make when shopping for financial services and products because the complex-
ity in fee disclosures reduces the benefits from searching. Data from the Survey of Consumer
Finances provide information about the shopping practices of Americans for financial
services. In 2010, based on a self-assessment of the effort made to shop for investments that
offered the best terms, 55% of families reported making a moderate effort, 23% reported
shopping a great deal for the best terms on investments, and 21% reported not shopping at
all (Bricker et al., 2013).

2. This article

We hypothesize that low fee providers of advisory services use less complexity in their fee
disclosures than high fee providers. Strategic complexity in fee disclosures makes it difficult
for people who are not financially sophisticated to compare fees across service providers. The
article develops analytical tools for analyzing complexity in fee disclosure. We do this by
using an expanded metric of complexity in language to evaluate the complexity of a sample
of disclosures, and then test this metric by asking college business students to compute
hypothetical fees and to rate the complexity of the fee disclosures.

216 L.A. Muller, J.A. Turner / Financial Services Review 25 (2016) 215–234



While fees are charged in a wide variety of ways, the traditional and most widely used
approach is to charge fees based on a percentage of the assets being managed. This approach
accounts for 85% of the fees received by advisory firms (Maxey, 2011).

We analyze fee disclosures for ongoing investment advisory or management services that
are available over the Internet. Data from the Survey of Consumer Finances indicate that the
Internet is used for information on investing by 80% of the population age 35 and younger
(Bricker et al., 2012). A number of major companies do not disclose their fees for financial
advisory services over the Internet, but instead require that the client contact them to obtain
that information. For this reason, this analysis of fee disclosures is for a truncated sample,
which is the sample of companies that disclose fees over the Internet.

The article begins by discussing related literature. It then examines the issue of complexity
in fee disclosures by adopting a methodology that is used to assess the grade level of writing
and expanding it to measure the complexity beyond simple sentence structure and word
choice. It applies this metric to all of the 10 largest advisory companies that provide fee
disclosures over the Internet, plus a sampling of other companies. Because many companies
do not disclose their fees over the Internet and because of the time consuming nature of the
process of finding the fees and then rating the associated disclosures, we have a small sample
of fee disclosures (n � 10). To validate this expanded metric, we survey a sample of 618
college business students for an independent rating of the disclosures. Finally, the article
presents concluding comments.

3. Literature review

This article on complexity in fee disclosures relates to several areas in the economic
literature.

In related work in computational linguistics, Loughran and McDonald (2013) examine the
readability or complexity of financial disclosures. More closely related to the hypotheses of
our article, Li (2008) argues that business managers attempt to hide the poor earnings
prospects of their firms by increasing the complexity of their written disclosures. He finds
that when annual reports are harder to read, good news they contain is more transitory and
bad news is more persistent in its effects on future earnings. Smith and Taffler (1992, 2000)
find that when a firm’s financial situation deteriorates, its accounting reports tend to become
more opaque. Older research finds issues of readability and complexity in material presented
in footnotes (Healy, 1977; Smith and Smith, 1971).

Using consumer search models, Ellison and Ellison (2009) use the term obfuscation to
describe marketing practices where firms make price comparisons complex or confusing.
Obfuscation can lead to reduced consumer learning about prices. Carlin (2009) and Wilson
(2010) also present models of strategic obfuscation where firms intentionally increase the
search costs of consumers. Carlin (2009) finds that when some consumers are unsophisti-
cated, firms can use complex pricing and price obfuscation to charge supracompetitive
prices. Ellison and Wolitsky (2011) present a search model where it is rational for firms to
increase the search costs of consumers. Prices and firm profits increase as search costs
increase. Price dispersion occurs in markets where consumers are differentially informed

217L.A. Muller, J.A. Turner / Financial Services Review 25 (2016) 215–234



(Ellison and Wolitsky 2011). Thus, price dispersion may occur in the market for financial
advisers because of heterogeneity in the level of financial sophistication of clients. Ellison
and Wolitzky (2012) find in some theoretical models that higher price mark-ups are asso-
ciated with greater obfuscation. They note that obfuscation is bad for consumers both in that
it raises the prices they pay and that it increases their search costs. Their model is based on
obfuscation increasing the amount of time it takes for consumers to discover prices. The
analysis in our article differs in that it focuses more on factors that make it difficult for
consumers to understand the fees that they would be charged, given that they have done the
search required to find that information.

This article, which focuses on advisory fee disclosures over the Internet, also relates to
literature on use of the Internet as a search mechanism. Ellison and Ellison (2009) argue that
the ease of search over the Internet has caused some firms to increase the obfuscation
concerning the disclosure of price and quality of products.

In an analysis of fee disclosures, the Government Accountability Office (U.S. Government
Accountability Office [GAO], 2013) reviewed fee disclosures online for 10 large providers
of Individual Retirement Accounts (IRAs). It finds that fees often are disclosed in ways that
make them difficult to understand. The lack of transparency involves a number of different
aspects of the disclosures, which may suggest a strategy of lack of transparency in fees. GAO
finds that fee information is generally scattered across the providers’ websites in ways that
make it difficult to find all the applicable fees. The results of the U.S. GAO (2013)
non-generalizable survey suggest that IRA fees are often disclosed in ways that imply that
fees are not important. The fees are sometimes located in difficult-to-find places. One
example is fees provided in the last section of a 49-page document. That section uses small
font type to disclose the information, another implicit message that the information is not
important. Fees are often located in footnotes in small font type. Footnotes generally are
viewed as being reserved for technical information that is not of interest to general readers.
Further obfuscating the fee disclosures, often the word “fee” is difficult to find.

Black et al. (2002) argue for comprehensive personal financial planning, but express
concern that such planning services may lead to less transparency in fees. Hung et al.
(2008) conduct a study for the Securities and Exchange Commission (SEC). Their main
purpose is to provide the SEC with a factual description of the investment advisory and
brokerage industries to assist the SEC in its evaluation of the legal and regulatory
environment concerning investment professionals. In their survey of investors, they find
that many investors find the disclosures provided by financial advisers to be difficult to
understand.

Starr (2010) examines issues relating to fees in 403(b) plans in the public and nonprofit
sectors. Mazzoli and Nicolini (2010) investigate determinants of the financial adviser’s
choice relating to the transparency versus opaqueness of pricing policy for financial advice
using Italian data. They do not include the level of fees as a factor explaining the degree of
opaqueness of the fee structure. They measure opaqueness by a binomial variable that is
categorized as opaque if the advisory fee is not charged as a separate element but is part of
other fees. Thus, they do not measure degrees of opaqueness for advisers who charge fees as
a percentage of assets, but consider all disclosures of that type of fees as transparent. Almost
all investment advisers charge fees based on assets under management, and thus their fee

218 L.A. Muller, J.A. Turner / Financial Services Review 25 (2016) 215–234



disclosures are considered to be transparent. Lachance and Tang (2012) find that trust is a
factor in the use of financial advisers. It may also be a factor in the use of financial advisers
who have opaque fee disclosures.

A study of consumers’ shopping strategies for Medicare Part D policies finds that insurers
profit when consumers fail to shop around for the best priced products (Ho, Hogan, and
Morton 2015). We argue in our article that some mutual fund providers encourage that result
by making information on fees difficult to find and difficult to understand.

4. Complexity in language

In 1998, the Securities and Exchange Commission adopted rules on plain English disclo-
sure in financial reports. The underlying argument behind the rules is that (1) firms could use
vague language to hide adverse information, and (2) the average investor may not understand
complex disclosures, which could reduce the efficiency of financial markets (Li, 2008).

A possible alternative motivation for the use of complex terminology in disclosures is that
it may impress some clients as to the sophistication of the service provider and the difficulty
of the subject. The client may view the apparent difficulty of the subject as the barrier to
comprehension, rather than viewing the barrier as being because of strategic complexity in
choice of language and expression.

Complexity can take the form of increasing the amount of time it takes to find fee
information. It can also take the form of increasing the amount of time it takes to understand
fee information, which would reduce the likelihood that fee information will be understood.
This section focuses on complexity that reduces the likelihood that fee information will be
understood.

Linguists have studied complexity in language and have developed empirical measures of
complexity. These measures rate written documents as to the grade level of education
required to understand them. These measures generally take into account sentence length and
the difficulty of the vocabulary used. Longer sentences are viewed as increasing the com-
plexity of language. Use of jargon or specialized terminology is one aspect of the difficulty
of the vocabulary used.

One commonly used measure for assessing complexity in the English language is the
Dale-Chall readability score (Dale and Chall, 1948). The formula that produces that score is

score � (0.1579 � PDW) � (0.0496 � ASL) � 3.6365 (1)

where PDW is percentage of difficult words and ASL is average sentence length.1 In using
the formula, 10% for PDW would be entered as 10. The score is then translated into a grade
level (see Table 1). Any word above the school grade 4 level (age 10) is rated as a difficult
word. The percentage of difficult words has a much larger effect on grade level of the writing
than sentence length. For example, a sentence can be 27 words long and still be rated at grade
level 4 if it has no difficult words, while if it has three difficult words, it would be rated at
a grade level 7–8.

This formula differs from some other readability formulas in that it assesses word
difficulty based on a subjective assessment rather than on counting number of letters or

219L.A. Muller, J.A. Turner / Financial Services Review 25 (2016) 215–234



syllables in words. The Dale-Chall approach uses a list of 3,000 words that are considered
the vocabulary of non-difficult words. A calculator is available on the Internet that checks
text against the list of 3,000 words (Readability Formulas, 2013b).2

A simple example demonstrates the use of the Dale-Chall approach for evaluating
financial disclosures. This analysis focuses on Fidelity Personalized Portfolios (Fidelity,
2013). Its fee disclosure is the following: “Gross annual advisory fee: Between 0.55% and
1.5% of eligible assets invested.” Using the Microsoft Word program to count the number of
words, this sentence has 12 words. That word counter appears to count the spaces between
strings of characters, so that 0.55% is counted as one word. Analyzing this disclosure as a
sentence, the Dale-Chall readability score is grade 16 and above.

This example points out a weakness of the application of the Dale-Chall readability
formula using their vocabulary list of 3,000 words for the purposes of analyzing financial
disclosures. The following words are not included in the list and, thus, are counted as difficult
words: annual, advisory, assets, invested. While these would be difficult words for many
fourth graders, they would presumably not be difficult words for pension participants reading
fee disclosures.

Thus, just as Loughran and McDonald (2013) conclude that the Gunning-Fog measure, an
alternative measure of language complexity, is not well-suited for analyzing financial
disclosures, we conclude that the 3,000 word vocabulary associated with the Dale-Chall
formula is too limited for the purposes of this article. We substitute instead a subjective
assessment of which words are difficult words for 401(k) participants. To measure readabil-
ity, the target audience needs to be taken into account.

The approach used in this article focuses on both difficult words and difficult concepts. For
example, the words “eligible” and “assets” are not assessed as difficult words for users of
financial advisory services, but the concept “eligible assets” is rated as a difficult or unclear
concept when it is not otherwise defined or explained. Similarly, while the word “fee” is not
a difficult concept, the phrase “gross fee” is rated as a difficult concept. In the above example
difficult concepts would be “gross fee” and “eligible assets.” Thus, for a sentence of 12
words, with four difficult words, applying the Dale-Chall formula with our assessment of
difficult words yields a rating of grades 13–15 (Table 1).

Long average sentence length is not the only aspect of sentence length that contributes to
complexity. Long total length of a disclosure can add to complexity. Furthermore, having

Table 1 Converting raw score to grade level using the Dale-Chall readability score

Raw score Grade level

0 to 4.9 4 or below
5.0 to 5.9 5 to 6
6.0 to 6.9 7 to 8
7.0 to 7.9 9 to 10
8.0 to 8.9 11 to 12
9.0 to 9.9 13–15 (college)
10.0 and above 16 and above

Source: Dale and Chall (1948).

220 L.A. Muller, J.A. Turner / Financial Services Review 25 (2016) 215–234



a single long sentence in a disclosure with multiple short sentences can add to complexity.
Both of these aspects of complexity are not measured by the Dale-Chall formula. For our
analysis, we stick, however, with the traditional approach of focusing on long average
sentence length.

Because we use a different assessment of difficult words than is used for the Dale-Chall
formula, we use the following score formula in the remainder of the article. This formula
uses the same parameter values as the traditional Dale-Chall formula for percentage difficult
words and average sentence length. The formula differs only in that we use a higher constant
because we use a more sophisticated vocabulary in measuring percentage difficult words.
The higher constant is used to calibrate the formula so that it yields a score and grade level
that is roughly comparable to what the Dale-Chall score would be if it used our vocabulary
of difficult words.

Adjusted Score � (0.1579 � PDW) � (0.0496 � ASL) � 6 (2)

In our formula, the percentage of difficult words is generally lower than for the Dale-Chall
formula because of using an expanded vocabulary of acceptable words. Appendix 1 contains
a partial list of the words found in disclosures that we rate as difficult. However, because of
our calibration of the constant term, Table 1 is still used to convert the score to a grade level.
The formula is specifically designed to rate complexity in documents that are designed for
a more sophisticated reader than one with a low level of education. Because of the constant
term, it cannot be used for rating documents for fourth graders.

In measuring the grade level of fee disclosures, a natural question is what should be the
grade level at which documents are written? According to one commentator on health
communications, “Health communications professionals generally recommend designing
adult targeted public education print materials for about a fifth- or sixth grade reading level,
to accommodate individuals who read at lower levels” (Sanner, 2003). The Joint Forum of
Financial Market Regulators (2004) in Canada has proposed that disclosures be written at a
fifth grade level.

5. Other elements of complexity

While we focus on complexity in language, the concept of complexity used in this article
is broader than just complexity in language. It is based on the idea that anything that
increases the difficulty that unsophisticated investors have in understanding how much they
pay in fees is part of the complexity in the presentation of fees. Complexity in fee disclosures
can be achieved in a number of different ways. Developing these complex approaches
imposes higher search and comprehension costs on users of financial advisory services.

This section discusses other ways in which fee disclosure can be made complex. A later
section provides examples of actual disclosures demonstrating these issues.

One strategy that financial advisers use is to provide an apparent disclosure that is simple
to understand, while the actual disclosure, often provided in a footnote, is difficult to

221L.A. Muller, J.A. Turner / Financial Services Review 25 (2016) 215–234



understand. The apparent disclosure may contain information on the level of fees, but
sometimes it does not. The apparent disclosure may be primarily a marketing tool. It differs
from the actual disclosure in that either it does not contain information on the level of fees
or the actual disclosure contains information that substantially modifies it. The actual
disclosure may incorporate the apparent disclosure, or it may be a separate disclosure that
contradicts in some respects the apparent disclosure.

The fee structure can be complex. It can require multiple calculations. It can involve
multiple fees for different aspects of a service. It can involve different ways that fees are
charged, with it being unclear as to which way would apply to the investor, or when there
is choice, which way would be most advantageous to the investor.

Some companies make it difficult to understand the asset base against which fees are being
charged. They may use terms such as “eligible assets,” “gross assets,” or “net assets.”

Complexity can be increased by placing disclosures in places that would signal that the
information is not important, such as footnotes. An example of complex disclosure is placing
the disclosure at the end of a lengthy segment of text, such as at the 28th line of footnotes.
Complexity can also be increased by disclosing fees in areas where they would not be
expected to be disclosed, such as under a different heading from where the apparent
disclosure occurs.

While most fee disclosures are short, some are lengthy. Sometimes length is because of
the addition of worked examples, or supplementary information to assist in understanding the
fees. We do not include worked examples and supplementary information as part of the basic
fee disclosure when we count sentence length.

Complexity can be increased by indicating a degree of uncertainty or ambiguity as to
actual fees, with use of the word “may” to indicate that actual fees may be greater or less.
Often this involves listing factors that may affect the level of fees, but not disclosing by how
much fees would be affected.

Some disclosures provide information that may mislead unsophisticated investors. For
example, Edelman (2013) provides the following statement: “you never pay any commis-
sions, brokerage fees or trading costs.” What this statement presumably means is that the
client does not pay these fees or costs as part of their advisory fee to Edelman. The client
does, of course, pay trading costs relating to the underlying investments. An unsophisticated
investor is not likely to see this distinction.

Complexity can be increased by making it more difficult to find fee information. One
measure of difficulty is the number of clicks it takes to get from the initial web page
describing advisory services to the page providing fee information.

With nondisclosure, it is impossible to obtain information as to level of advisory fees
because that information is not provided. A number of major financial service providers do
not disclose information about their fees on the Internet, but instead require that a potential
client talk to an adviser.

Complexity can involve not disclosing information about the fees charged by the invest-
ment products that the adviser recommends. In practice, rarely do disclosures provide
information as to the range of fees or typical fees of the investment products they recom-
mend. While that is not an aspect of the compensation of the adviser, it is an aspect of the
cost to the client.

222 L.A. Muller, J.A. Turner / Financial Services Review 25 (2016) 215–234



6. Empirical analysis rating fee disclosures

In this section, we examine the fee disclosures of a small number of firms. Because we
individually code the complexity of the fee disclosures, the sample size is small. This
empirical analysis is intended to demonstrate the use of the analytical tools we develop, but
also to provide suggestive evidence as to the relationship between level of complexity of fee
disclosure and level of fees. We argue that complexity is a multidimensional (or complex)
strategy.

Our small sample of fee disclosures is limited in several ways. First, it is limited to firms
that disclose their fees over the Internet. Second, it is limited to firms that do not also receive
compensation through commissions. Part of the fee paid by the participant may be hidden in
the form of higher expense ratios on the investments to offset the commissions the mutual
fund companies pay directly to the adviser. Third, the sample is limited to firms with a
minimum asset level for advice of no more than $500,000. We pick that figure because it is
the minimum used by a number of companies. Arguably, investors with higher amounts
would tend to be more sophisticated and thus more likely to be able to understand complex
fee disclosures. Furthermore, a limit above that amount would exclude most people.

In rating grade level of the text of the disclosures, we apply the adjusted Dale-Chall
readability formula (Eq. 2), but instead of using their vocabulary list, we apply our own
subjective assessment as to which words would be considered difficult. Computer analyses
of texts can have difficulty determining sentence length because not every period denotes the
end of a sentence. That problem is not an issue in the analysis in this article, where we
visually count sentence length.

The 10 largest mutual fund companies in the United States are Vanguard, American,
Fidelity, T. Rowe Price, PIMCO, Franklin Templeton, BlackRock, Oppenheimer, JP Mor-
gan, and Columbia (Roth, 2012). The top three companies—Vanguard, American, and
Fidelity—controlled $2.6 trillion in assets in 2008 (Vohwinkle, 2008). We examine fee
disclosures for all of these companies. Not all of these companies, however, provide
on-going investment advisory services. Oppenheimer, for example, does not appear to
provide an advisory service for ongoing financial advice. Of these 10 mutual fund compa-
nies, we were able to find on their Internet websites information about fees for ongoing
financial management or advice for three: Vanguard, Fidelity, and T. Rowe Price. Perhaps
as important as fee disclosure is fee nondisclosure. While it appears that some of the
companies do not provide ongoing advisory services for individual clients, some companies
describe those services but do not disclose their fees at their website. For example, a search
of the website for BlackRock Managed Accounts did not yield any information concerning
fees (BlackRock, 2013).

Because of our inability to find information on the Internet about fees for most of the
largest companies, we supplement the information for those companies that is available with
information from a convenience sample of companies that do provide information about their
fees on the Internet. Because the sample is small, and it is a nonrandom sample, the sample
cannot be used to statistically test our hypothesis. The companies in the sample are included
based on their providing a fee disclosure on the Internet. The companies range in size from

223L.A. Muller, J.A. Turner / Financial Services Review 25 (2016) 215–234



very large ones to very small ones. They include companies that also sell investment products
and companies that only provide advisory services.

Table 2 provides some descriptive information about fee disclosure. Some companies
provide information about fees in the apparent fee disclosure, but do not actually disclose the
level of fees. We have not included those companies in this analysis. Many companies do not
provide information about their fees at their websites, and those are not been included. The
companies in Table 2 are thus a convenience sample of companies that provide actual fee
disclosures on their websites.

A general pattern can be seen among the small number of advisor firms in Table 2. Firms
charging higher fees are more likely to provide a simplified apparent disclosure along with
an actual disclosure. For those companies, the actual disclosure tends have longer sentences
and more total words, both indicators of complexity.

Table 3 rates both the apparent fee disclosure and the actual fee disclosure in terms of
grade level. The apparent fee disclosure is at a considerably lower grade level than the actual
fee disclosure when both are provided. The main finding from the sample in Table 3 is that
the grade level of the actual disclosure is considerably higher in high fee companies than in
low fee companies. Thus, while complexity is presumably a strategy, high fee providers can
be identified solely based on the complexity of the language they use.

An alternative hypothesis is that high-fee providers have more complex disclosures
because they provide a more complex range of services. While that may be true for some

Table 2 Description of fee disclosures

Adviser company Apparent
disclosure
number of
words (mean
per sentence)

Actual
disclosure
number of
length (mean
per sentence)

Fees on first dollar (minimum
investment)

Mean Total Mean Total

1. Edelman Financial Services
(2013)

5 5 58 58 2.0%

2. Edward Jones (2013) – – 28 309 1.8% ($500,000)
3. Fidelity (2013) 12 12 13 65 1.5% ($200,000)
4. West Financial Services 11 11 25 76 1.25% plus $250 an hour for financial

planning consulting
5. Summit Financial Services (2013) – – 16 16 1.0% ($500,000)
6. FBB Capital Partners – – 17 17 1.0% ($500,000)
7. Vanguard (2013) – – 13 80 0.9% ($500,000)
8. Zero Commission Portfolio
(2013)

– – 17 17 $149 a month

9. BloombergBlack (2013) – – 11 11 $100 a month
10. NestWise20 (2013) – – 8 8 $250 initial fee and $575 a year

Source: Authors’ compilation.
Note: The average sentence length is in some cases an approximation because of the use of headings and tables.

The total length of disclosures does not include a worked example, if one is provided. Providing an example adds
to the length, but it would not be an aspect of complexity in presentation. The level of fees is the fee charged on
the first dollar of investments. When a range of fees is specified without other information, the highest fee is
reported.

224 L.A. Muller, J.A. Turner / Financial Services Review 25 (2016) 215–234



types of financial advisers, that hypothesis is rejected in our sample by examining the
services provided. In all cases, the only service provided is advice for managing investments.

A further hypothesis is that any person doing comparison shopping for a financial
management firm would tend to choose one with low fees and an understandable disclosure.
Similarly, anyone initially choosing an adviser with high fees, would tend to change to a
lower fee adviser over time. Those hypotheses assume rational, well-informed behavior, and
would argue against the existence of high-fee and low-fee providers providing similar
services, which is actually what is observed in the market.

7. Survey of college students: validation of our rating tool

In the previous section, we note several dimensions of complexity—complexity in
language, complexity in computations, complexity in finding relevant information. In the
remainder of the article, we focus on the dimension of the complexity of language. We make
that choice because we believe that dimension is an important dimension of complexity, and
also because we want to narrow the scope of our research so as to be able to focus on one
aspect of the issue.

To validate our rating tool as a measure of the difficulty of understanding fee disclosures,
we surveyed 618 students in the Seidman College of Business at Grand Valley State
University, a midsize university in Allendale, MI. While all of the students were taking
business classes, and most of the students in the sample are business majors, 27% of the
respondents are not business majors, including 16% who are studying mathematics intensive
majors, such as engineering, but are not business majors (Table 4). While the sample of
students is selective in some respects, it is less selective than the sample used by Choi,
Laibson, and Madrian (2010), where the sample consisted largely of students from Harvard
and Wharton, and the SAT scores averaged in the 99th and 98th percentiles, respectively.
The survey was voluntary and administered online by the school Blackboard site in October

Table 3 Rating fee disclosures

Adviser company Grade level Fee on first dollar

Apparent
disclosure

Actual
disclosure

1. Edelman 9–10 16 2.0%
2. Edward Jones (2013) – 13–15 1.8%
3. Fidelity (2013) 9–10 13–15 1.5%
4. West Financial Services 7–8 11–12 1.25% plus $250 an hour for financial

planning consulting
5. Summit Financial Service (2013) – 7–8 1.0%
6. FBB Capital Partners – 7–8 1.0%
7. Vanguard (2013) – 9–10 0.9%
8. Zero Commission Portfolio (2013) – 7–8 $145 a month
9. BloombergBlack (2013) – 7–8 $100 a month

10. NestWise (2013) – 7–8 $250 initial fee, $575 annual fee

Source: Authors’ compilation, based on a portfolio of $425,000.

225L.A. Muller, J.A. Turner / Financial Services Review 25 (2016) 215–234



2014. The goal was to present the students with two actual fee disclosures that we have rated:
one that we rated as “simple” and one that we rated as “complex,” using our adjusted
readability formula (Eq. 2). We did not identify in the survey which disclosure was simple
and which was complex. The students would then compute the fee charged, as well as
rate the complexity of the disclosure on a scale from 1 to 10. The full survey is found
in Appendix 2.

Summary statistics for the sample are found in Table 4. A little over half of respondents
are male, and 73% of the students are in some type of business major. Nonbusiness majors
comprise 27% of the sample, with 11% having a non-math intensive major such as education
or sociology, and 16% having a math-intensive major such as engineering or statistics.
Additionally, 88% are age 25 or under, with the remaining 12% being nontraditional students
or individuals pursuing their MBA. Ninety-six percent of respondents list English as their
first language.

Some characteristics of our sample may bias the results either upwards or downwards with
respect to the population of those reading financial disclosures. The literature suggests that
age is positively related to financial knowledge, so given that most of the respondents are
under age 25, the age of the respondents in our sample would bias the results downward,
meaning they would be more likely to rate a disclosure as complex. For example, Lusardi and
Mitchell (2011) find that financial literacy tends to have an inverted-U pattern with respect
to age, being lowest among the young and the old. However, the literature also shows that
individuals with business degrees make more sophisticated financial decisions (Allgood et
al., 2011), biasing the results upwards. Given that we do not have sufficient information to
determine the degree of bias either way, we should keep this in mind when extrapolating the
results to the general population.

Table 4 Summary statistics for survey sample

Characteristic Percentage of sample

Male 54
Female 46
Age 25 and under 88
Age over 25 12
English is first language 96
English is not first language 4
Non-business majors 27

Non-math intensive 11
Math intensive 16

Business majors 73
Accounting 20
Economics 5
Finance 12
Management 14
Marketing 14
General business 6
MBA 2

Source: Authors’ tabulations, based on a survey sample of 618 students.
Note: Non-business, non-math intensive majors would include, for example, education or sociology majors.

Non-business, math intensive majors would include, for example, engineering and statistics majors.

226 L.A. Muller, J.A. Turner / Financial Services Review 25 (2016) 215–234



In Table 5 we present the answers the students gave when asked to compute the fee from
both the simple and the complex disclosure (questions 5a and 6a, respectively; see Appendix
2). The results clearly show that the students had a more difficult time with the complex
disclosure. While half got the correct answer to the simple disclosure, only 15% computed
the fee correctly for the complex one. In addition, while only 24% answered “don’t know”
for the simple disclosure, 45% gave this answer for the complex one.

The results in Table 6 confirm the findings in Table 5—the disclosure we rate as complex
is more difficult than the one we rate as simple. While 40% of students rated the simple
disclosure as very easy (1–3 on a scale from 1 to 10), only 24% rated the complex disclosure
as such. Furthermore, while approximately 10% rated the simple disclosure as a 9 or 10
(most difficult), 18% rated the complex one as such. Thus, our findings are consistent with
earlier studies finding heterogeneity in the ability of people to deal with complexity in
financial issues.

8. Logit analysis

We also ran logit regressions to examine what characteristics of the student were corre-
lated with a correct fee calculation for the simple fee disclosure and for the complex

Table 5 Student responses to questions calculating the amount of fees, in percentages

Response Simple disclosure (percent) Complex disclosure (percent)

Can’t tell from information provided 14.8 10.2
Answer given, correct 49.8 15.0
Answer given, incorrect 11.3 29.8
Don’t know 24.1 45.0
Total 100.0 100.0
Number of responses 576 618

Source: Authors’ tabulations from survey. Simple disclosure is question 5a; complex disclosure is question 6a
(see Appendix 2).

Table 6 Rating the difficulty of financial disclosures

Student rating of the complexity
of the fee disclosure

Simple disclosure Complex disclosure

Percent Cumulative
percent

Percent Cumulative
percentage

1–3 40.4 40.4 23.5 23.5
4–5 24.1 64.5 20.5 44.0
6–8 25.7 90.2 38.3 82.3
9–10 9.8 100.0 17.7 100.0
Total 100.0 – 100.0 –
Number of responses 618 – 611 –

Source: Authors’ tabulations from survey. Ratings range from 1 (easiest) to 10 (most difficult). Simple
disclosure is question 5b; Complex disclosure is question 6b (see Appendix 2).

227L.A. Muller, J.A. Turner / Financial Services Review 25 (2016) 215–234



disclosure. This analysis provides insights as to the effect of complexity on the ability to
correctly determine a price, which is the fee charged for financial advice. The marginal
effects are reported in Table 7 for both regressions.

In both models, age and whether the student is a native English speaker are both
statistically insignificant. For the simple fee disclosure, men and women are equally likely
to provide the correct answer. However, for the complex disclosure, men are 8% more likely
to correctly compute the fee than were women. Other studies show that women tend to have
lower financial literacy than men (Lusardi and Mitchell, 2008).

We are especially interested in whether the student’s major has an effect on whether he
or she answered the question correctly, as courses related to financial decision making are
one of the ways a person can develop financial sophistication. Hence course major can serve
as a proxy of financial sophistication. Accounting and finance majors are the most likely
group to have specifically studied fee disclosure. In addition, MBA students are also exposed
to higher-level accounting and finance material that may aid them in answering the questions.
MBA students tend to be older and have had experience in financial decision making, so they
may be even more financially sophisticated than the undergraduates. However, since we are
not able to disentangle the effects of this aspect of sophistication from the knowledge they
gain in the MBA program, we cannot assume that the MBA variable is a proxy of only
coursework. The remaining business majors—marketing, management, and general busi-
ness—have had exposure to business terminology and may have had accounting or finance
classes, but not studied fee-disclosure type problems extensively. Economics and nonbusi-
ness majors with math-intensive concentrations (e.g., engineering or statistics) may not have
had the exposure to business terms, but may have the mathematical capability to compute the
fees. The last group—and omitted category in the regression—is students who have a
nonbusiness, non-math intensive major (e.g., elementary education or sociology). These

Table 7 Logit analysis

Characteristic Low fee–correct answer High fee–correct answer

Age over 25 .04 (.07) .06 (.05)
Male .01 (.05) .08 (.03)**
Native English speaker .19 (.10) .03 (.05)
Major: Accounting .30 (.07)** .33 (.15)*

Economics .14 (.11) .07 (.14)
Finance .25 (.08)** .23 (.15)
Management .09 (.09) .11 (.12)
Marketing .02 (.10) .13 (.13)
General business .07 (.10) �.01 (.10)
MBA .29 (.11)* .51 (.21)**
Non-business/math .18 (.10) .17 (.14)

N 575 617

Source: Authors’ calculations from survey.
Note: The estimates above are marginal effects. The category Non-business/math refers to non-business majors

whose major include a significant amount of math (e.g., engineering). The omitted major category is a
non-business major in a concentration with little to no math (e.g., elementary education, sociology).

*p � 0.05, **p � 0.01.

228 L.A. Muller, J.A. Turner / Financial Services Review 25 (2016) 215–234



students would be the least likely to be able to correctly compute the fees, as their exposure
to business terminology and mathematical computations is most likely minimal.

For the most part, the marginal effects with respect to major are consistent with what is
learned in the classroom. For the simple disclosure, accounting and finance majors are 30%
and 25% more likely, respectively, to compute the correct answer than are students with a
nonbusiness, non-math intensive major. MBA students are also more likely to compute the
correct answer.

Generally, it might be assumed that most people would understand the simple disclosure,
while people who were more sophisticated in terms of financial literacy would be more likely
to understand the complex disclosure. However, while there is still a large, statistically
significant effect for MBA students, for accounting and finance majors the effect for the
complex disclosure was less statistically significant or insignificant, respectively. So there is
some evidence that financial sophistication helps, but helps to a lesser extent with the more
complex fee disclosures.

9. Policy analysis

Clearly, some fee disclosures are written at a high grade level. Whether that is done to
purposely obscure the disclosure, as argued in this article, or for other reasons, a public
policy response could be to have readability requirements for fee disclosures. American Plain
English Statutes in some states require that legal text be written below a specified grade level
(Stumpff, 2013).

10. Conclusions

While much attention has been given to financial education for pension participants, less
attention has been given to the quality of information they receive from financial service
providers. This article hypothesizes that some financial advisers take advantage of the lack
of sophistication of their clients by disclosing fee information in complex ways.

Complexity is a multidimensional strategy that some investment management or advisory
companies use to make it difficult, if not impossible, to compare advisory services on the
basis of fees, and to determine what level of fees a prospective investor would be paying. We
note several dimensions of that strategy—complexity in language, complexity in computa-
tions, complexity in finding relevant information—but in the end focus on the dimension of
the complexity of language. We make that choice because we believe that dimension is an
important dimension of complexity, and also because we wanted to narrow the scope of our
research so as to be able to focus on one aspect of the issue. The strategy takes advantage
of the lack of financial sophistication of many investors, but in some instances even
sophisticated investors may not be able to make comparisons or obtain accurate fee infor-
mation based on the disclosures provided on the Internet. This strategy presumably reduces
competition in the market for advisory services and permits service providers to charge
higher fees, raising their profits.

229L.A. Muller, J.A. Turner / Financial Services Review 25 (2016) 215–234



Our article develops a hypothesis concerning strategic complexity in fee disclosures,
develops a measure of complexity in fee disclosures, collects a small sample of fee disclo-
sures, notes the simple correlation between complexity in fee disclosure according to our
measure and level of fees, and collects a substantial sample to validate our measure of
complexity of fee disclosures. Our nonscientific sample of 10 fee disclosures is only
suggestive of the relationship between complexity in fee disclosures and level of fees. Our
larger sample validates our measure of complexity. It indicates that people who are more
financially literate (based on college studies) are better able to understand complex fee
disclosures, but that even people with the presumption of a relatively high degree of financial
literacy are not all able to decipher complex fee disclosures.

This article develops analytical tools for measuring complexity in fee disclosures. It
presents suggestive evidence, based on a small sample (10 advisory companies), that
higher-fee advisers use more complex disclosures than lower-fee advisers. It also finds that
complexity in disclosure in one dimension is generally accompanied by complexity in
disclosure in other dimensions, suggesting use of strategic complexity in disclosures. In
particular, higher-fee advisers tend to use more complex language and longer sentences,
present important information in footnotes, and have more complex fee structures.
However, it is possible to identify high fee providers solely on the basis of the
complexity of the language they use. The article rejects by the use of counterexamples
the argument that complexity is inherent and that simple fee disclosures involve a
tradeoff against completeness.

While financial education has been recommended by a number of commentators as the
solution to dealing with lack of financial sophistication, with strategic complexity in fee
disclosure, financial education may be of little use. For some fee disclosures, the complexity
of language is beyond that that would reasonably be understood.

The issue of strategic complexity in fee disclosures may exist in other aspects of financial
services. For example, future studies could explore this issue concerning credit card fees, fees
for financial products, fees for mortgages and other loans, fees for insurance products, and
banking fees. The structure of products in all these areas has become increasingly complex.
Further research could explore those areas and test whether in those areas higher-fee
providers also tend to use more complex fee disclosures.

Notes

1 A different version of the formula only includes the constant 3.6365 if the PDW is five
percent or greater, which is generally for reading at grade 3 or lower.

2 We examined two other options for rating complexity of language, and found both to
be inappropriate for evaluating fee disclosures. The Fog index, sometimes called the
Gunning-Fog index, developed by Robert Gunning (1952), bases readability on
sentence length and the proportion of words with more than two syllables. In an
analysis of the use of that index for measuring the readability of financial disclosures

230 L.A. Muller, J.A. Turner / Financial Services Review 25 (2016) 215–234



Loughran and McDonald (2013) conclude that it is not well suited for measuring the
complexity of financial disclosures because words with two or more syllables, such as
“corporation,” are common in financial disclosures. The Flesch Reading Ease test is
also based on sentence length and the average number of syllables per word (Read-
ability Formulas, 2013a).

Acknowledgment

We have received helpful comments from Bernard Casey, Vijay Gondhalekar, Olivia
Mitchell, David McCarthy, an anonymous reviewer, and participants in the 21st Annual
Conference of Superannuation Researchers in Sydney, Australia; the conference on Pension
Reforms in Post-Socialist and Other Countries in Poznan, Poland, the 2013 Conference of the
European Network for Research on Supplementary Pensions (ENRSP) in Münster, Germany,
and the 2015 Western Economic Association International conference in Honolulu, Hawaii.

Appendix 1: Difficult words and concepts in financial disclosures

The following is a list of words or concepts that appear in financial advisory fee
disclosures that are subjectively rated as difficult in terms of their effect on understanding the
fees that a person would pay.

eligible assets, gross assets, net assets, gross fees, net fees, 12b-1 fees, transfer fees, SEC
fees, wrap fees, net assets, load, may.

Appendix 2: Survey

Understanding fees for financial products and services

Please answer the following questions. you may use a calculator.

Demographic questions

1. Please indicate your gender M or F
2. What is your age?
3. Is English your native language? Y or N
4. What is your major/intended major?

Fee disclosures for financial advice

5. “Start with a free 60-day trial, then pay $100 a month.”

231L.A. Muller, J.A. Turner / Financial Services Review 25 (2016) 215–234



a. Based on that fee structure, how much would you pay the first year for an investment
account that averaged $500,000 over the year? Check one.

Can’t determine from the information .
The amount would be .
I don’t know
b. On a scale of 1 to 10, with 1 being easy and 10 being impossible, how difficult was it

to answer question 5a?
6. “We feature a single annual fee, calculated and debited quarterly from your account. It

appears directly on your statement. The annual fee schedule is shown below.”

Amount invested Fee

First $150,000 2.00%
Next $250,000 1.65%
Next $350,000 1.25%
Next $250,000 1.00%
Next $2 million 0.75%
Next $7 million 0.60%
Next $15 million 0.50%
Above $25 million Negotiable

a. Based on that fee structure, how much would you pay the first year for an investment
account that averaged $500,000 over the year? For simplicity, assume the account balance
is $500,000 each quarter. Check one.

Can’t determine from the information .
The amount would be .
Don’t know
b. On a scale of 1 to 10, with 1 being easy and 10 being impossible, how difficult was it

to answer question 6a?

References

Abeler, J., & Jäger, S. (2015). Complex tax incentives. American Economic Journal: Economic Policy, 7, 1–28.
Allgood, S., Bosshardt, W., van der Klaauw, W., & Watts, M. (2011). Economics coursework and long-term

behavior and experiences of college graduates in labor markets and personal finance. Economic Inquiry, 49,
771–794.

Black, K., Ciccotello, C. S., & Skipper, H. D. (2002). Issues in comprehensive personal financial planning.
Financial Services Review, 11, 1–9.

BlackRock. (2013). What are managed accounts? http://www2.blackrock.com/us/individual-investors/products-
performance/managed-accounts.

Bricker, J., Kennickel, A., Moore, K., & Sabelhaus, J. (2012). Changes in U.S. family finances from 2007 to 2010:
Evidence from the Survey of Consumer Finances. Federal Reserve Bulletin, 98.

Carlin, B. (2009). Strategic price complexity in retail financial markets. Journal of Financial Economics, 91,
278–87.

Choi, J.J., Laibson, D., & Madrian, C. (2010). Why does the law of one price fail? An experiment using index
mutual funds. Journal of Financial Studies, 23, 1405–1432.

Dale, E., & Chall, J. (1948). A formula for predicting readability. Educational Research Bulletin, 11–20.

232 L.A. Muller, J.A. Turner / Financial Services Review 25 (2016) 215–234



Edelman Financial Services. (2013). Invest with Edelman Financial (available at http://www.ricedelman.com/
cs/invest_with_edelman/investment_management_how_emap_works).

Ellison, G., & Ellison, S. (2009). Search, obfuscation, and price elasticities on the Internet. Econometrica, 77,
427–452.

Ellison, G., & Wolitzky, A. (2012). A search cost model of obfuscation. Rand Journal of Economics, 43,
417–441.

Fidelity. (2013). Fidelity managed accounts (available at https://www.fidelity.com/managed-accounts/services).
Franklin Templeton Investments. (2013). Separately managed accounts (available at https://www.franklintempleton.com/

retail/pages/generic_content/prog_serv/ftpa/separately_mgd_accts.jsf).
Gunning, R. (1952). The technique of clear writing. New York: McGraw-Hill.
Healy, P. (1977). Can you understand the footnotes to financial statements? Accountants Journal, 219–222.
Ho, K., Hogan, J., & Morton, F. (2015). The impact of consumer inattention on insurer pricing in the Medicare

Part D Program. NBER Working Paper No. 21028.
Hung, A. A., Clancy, N., Dominitz, J., Talley, E., Berrebi, C., & Suvankulov, F. (2008). Investor and industry

perspectives on investment advisors and broker-dealers. LRN-RAND Center for Corporate Ethics, Law, and
Governance within the RAND Institute for Civil Justice and commissioned by the U.S. Securities and
Exchange Commission (SEC) (available at https://www.sec.gov/news/press/2008/2008–1_randiabdreport.
pdf).

Joint Forum of Financial Market Regulators (Canada). (2004). Guidelines for capital accumulation plans, (May
28).

Lachance, M.-E., & Tang, N. (2012). Financial advice and trust. Financial Services Review, 21, 209–226.
Li, F. (2008). Annual report readability, current earnings, and earnings persistence. Journal of Accounting and

Economics, 45, 221–247.
Loughran, T. & McDonald, B. (2013). Measuring readability in financial disclosures (available at http://

papers.ssrn.com/sol3/papers.cfm?abstract_id�1920411).
Lusardi, A., & Mitchell, O. (2007). Baby boomer retirement security: The roles of planning, financial literacy, and

housing wealth. Journal of Monetary Economics, 54, 205–224.
Lusardi, A., & Mitchell, O. (2008). Planning and financial literacy: How do women fare? NBER Working Paper

No. 13750, (January).
Lusardi, A., & Mitchell, O. (2011). Financial literacy around the world: An overview. NBER Working Paper No.

17107 (available at https://core.ac.uk/download/files/153/6358633.pdf).
Maxey, D. (2011). How to pay your financial adviser, The Wall Street Journal, December 12 (available at

http://online.wsj.com/article/SB10001424052970204554204577024152103830414.html).
Mazzoli, C., & Nicolini, G. (2010). To fee or not to fee: Pricing policies in financial counseling. Financial

Services Review, 19, 307–322 (available at https://www.researchgate.net/profile/Gianni_Nicolini/publication/
46536799_To_fee_or_not_to_fee_pricing_models_in_financial_counselling/links/5441490f0cf2a6a049a
56e3e.pdf).

McCarthy, D., & Turner, J.(2000). Pension education: Does it help? Does it matter? Benefits Quarterly, 16,
64–72.

Readability Formulas. (2013a). The Flesch grade level readability formula (available at http://www.readability-
formulas.com/flesch-grade-level-readability-formula.php).

Readability Formulas. (2013b). Free Dale-Chall readability formula with word list—Original and Revised
Versions (available at http://www.readabilityformulas.com/free-dale-chall-test.php).

Roth, A. (2012). Morningstar grades ten largest mutual fund families. Moneywatch, (July 9).
Sanner, B. (2003). Are your written materials missing the mark? Journal on Active Aging, (July-August), 18–24.
Smith, M., & Taffler, R. J.(1992). The chairperson’s statement and financial performance. Accounting and

Finance, 32, 75–90.
Smith, M., & Taffler, R. J. (2000). The chairperson’s statement: A content analysis of discretionary narrative

disclosures. Accounting, Auditing & Accountability Journal, 13, 624–647.
Smith, J. E., & Smith, N. P.(1971). Readability: A measure of the performance of the communication function

of financial reporting. Accounting Review, 46, 552–561.

233L.A. Muller, J.A. Turner / Financial Services Review 25 (2016) 215–234



Starr, B.(2010). Willingness to pay: Understanding 403(b) fees. Financial Services Review, 19, 125–144.
Stumpff, A. (2013). Synthetic common law: Toward simpler legal rules. Paper presented at conference at

University of Michigan.
T. Rowe Price. (2013). Advisory planning services (available at http://individual.troweprice.com/public/Retail/

Products-&-Services/Advisory-Planning-Services).
U.S. Government Accountability Office (GAO). (2013). 401(k) plans: Labor and IRS could improve the rollover

process for participants. GAO-13–30.
Vanguard. (2013). Vanguard asset management services fees (available at https://investor.vanguard.com/what-

we-offer/investing-help/partner-with-me-for-ongoing-help?Link�facet).
Vohwinkle, J. (2008). List of 50 largest mutual fund companies ranked by assets for 2008. Generation X Finance

(available at http://genxfinance.com/list-of-50-largest-mutual-fund-companies-ranked-by-assets-for-2008/).
Wells Fargo. (2013). Wells Fargo advisors’ services and fees (available at https://www.wellsfargo.com/beyond

today/tools-checklists/advisorinfo/advisorfees).
Wilson, C. (2010). Ordered search and equilibrium obfuscation. International Journal of Industrial Organization,

28, 496–506.

234 L.A. Muller, J.A. Turner / Financial Services Review 25 (2016) 215–234