FINANCIAL SERVICES AND FINANCIAL INSTITUTIONS:
VALUE CREATION IN THEORY AND PRACTICE
J. Kimball Dietrich
Optimizing Financial Institution Performance
Optimizing financial institution performance consists of getting financial service
employees to work efficiently with each other and with the economic and financial resources
with which they are entrusted. The topics considered in this chapter are broadly concerned with
organizing and paying financial institution employees and are concerned with the following
What approaches and issues are associated with compensation in the financial services
What considerations enter into how financial service firms organize their activities and
what are the connections between organizational form and value production?
How do multidivisonal financial firms allocate costs and profits?
What concerns should management of financial firms focus on as these firms grow and
adapt to environmental changes?
These topics have been addressed by economists as discussed below but are usually not central to
economic analysis. These subjects are less susceptible to rigorous analysis than other topics we
have addressed in this book but their overwhelming importance mandates that these issues be
highlighted in this book for future managers of financial service firms.
24.1 Employee Compensation
The core of value creation in financial service firms is good people, workers who can
communicate with customers, process and analyze complex information, negotiate deals and
deliver quality products and services. Financial institutions are usually complex organizations --
multidivisional firms like Fifth Third Bancorporation or Country Wide Financial or often large
holding companies like BancOne or Travellers with many separate corporate entities offering
different financial services in different markets. How can workers' personal interests be brought
into line with those of their employers in large organizations?
The problem of motivating workers can be viewed as a principal-agent problem as
introduced in Chapter 6 and explored extensively in Part II which analyzed sources of value in
providing financial services. In that discussion, principals and agents were typically financial
service firms and their customers involved in some relationship like a credit arrangement or
insurance policy. In discussing financial service firm workers, the financial service firm as
employer is the principal and the worker is the agent whose actions can harm or help the firm.
An extensive literature has evolved examining the workers' relations to employers1.
Problems between financial service firm employers and their workers are similar in many
ways to manufacturing firms and very different in others. In all firms, workers' interests are to
have high pay and make a low level of effort. Their employers' desire to pay workers little and
extract maximum effort. Employers cannot tell how much effort workers put out relative to their
capacity. Economic analysis of employment contracts examines compensation schemes which
minimize the cost to employers of their inability to control or perfectly monitor their workers'
efforts because of information asymmetries.
In analysis of the simplest case2, workers are characterized as being like tenant farmers
who can vary their effort in working the land and their resulting output is observable. The
question addressed is how the employer or landlord owning the land and the worker or tenant-
farmer can enter an efficient compensation scheme in the sense that workers put out maximum
effort. Analysis of this simple case demonstrates that the optimal compensation arrangement is
for the landlord to charge the farmer a fixed rent and allow the farmer to keep all the production
from the land. Since the worker or tenant farmer claims all output above a fixed amount, the
worker is a residual claimant on production. Workers have the incentive to maximize effort and
expected income equal to expected production minus the rent. In the simple case, employers pay
workers the value of their output above a fixed amount.
Incentive contracts between principals and agents in the employment context must be
altered to accommodate variations from the simplest case. One important potential difference
between workers and employers are risk preferences reflected in differing tolerance to variations
in income. Workers supporting families and depending exclusively on labor income to live may
be more risk averse than employers who are wealthy owners or managers of corporations owned
by diversified investors not exposed to firm specific risk. Efficient compensation contracts when
employee are risk averse typically include risk sharing by the employer in exchange for increased
income security for workers. In the case of the tenant farmer, the landlord shares output
variations with tenant farmers, becoming residual claimants to output like the farmer. Workers
do not receive all the value of their output above a fixed amount. This compensation
arrangement causes an efficiency loss since the farmer shares the benefits of effort with the
landlord and will be less willing to put out maximum effort in production.
Other complications arise when the worker (or tenant farmer) cannot be perfectly
monitored or when there are information asymmetries. Imperfect monitoring occurs when output
or factors like weather cannot be observed and related to production, becoming confused with the
effect of the worker's efforts. Information assymmetries occur when workers know more about
production conditions as when farmers know more than landlords about growing conditions and
the productivity of land. As in the case of risk aversion, in these departures from the simplest
See Sappington (1991) for a survey of compensation schemes
or Milgrom and Roberts (1992) for a complete development of many
of the issues in this chapter.
This treatment loosely follows Sappington (1991).
case optimal compensation arrangements for workers involve sharing output and distributing the
risk of output variations between worker and employers.
This discussion demonstrates than in departures from the assumptions of the simplest
case, optimal compensation requires sharing output or profits and distributing risks of output
variation between workers and employers. While sharing output reduces workers' incentives to
put out maximum effort because they gain only part of the payoff, these contracts are the most
efficient arrangements and align the interests of employers and workers as much as possible. In
view of apparent risk aversion among workers in financial services and the likelihood of
imperfect monitoring and output measurement by employers, it is surprising how few
compensation arrangements at lower levels of financial service firms are characterized by sharing
value of production or profits. Most financial service workers are paid by the hour or on an
Given the labor intensivity and the value of human capital in financial services,
compensation arrangements are vitally important to induce optimal performance of workers in
financial service firms. Optimal employee compensation is at the heart of value production in
the future evolution of this industry. Nonetheless, compensation of financial service firms differ
widely among industry segments. Some firms are creative and innovative in paying workers
while others cling to standard employer-employee arrangements. Compensation specialists are in
great demand in the financial service industry.
To place employee compensation into a framework, we list and define standard
compensation schemes observed in the financial services and other industries. We then
characterize some distinguishing characteristics of financial service employment and assess the
importance of the uniqueness of value creation in this industry in determing optimal employee
Workers' pay can be determined seven different ways: (1) straight time (typically hourly);
(2) salary (usually annual); (3) piece work; (4) participation contracts; (5) bonus systems; (6)
tournaments; and, finally, (7) equity participation. These compensation schemes are briefly
defined and illustrated in Table 24-1. We will use this breakdown in analyzing compensation in
financial services in the following discussion.
Each of the payment schemes shown in Table 24-1 is used in some segments of the
financial services industry for some jobs. Often hybrid payment systems are used, for example
combining salary and bonus systems. However, straight time and salary compensation schemes
dominate worker pay in most financial service firms. In view of the simple standard analysis
discussed above, straight time and salary compensation methods may be the least desirable
compensation schemes from the viewpoint of economic efficiency in financial services. These
pay methods do not address principal-agent problems associated with employees of financial
Straight time pay means simply punching a clock and being paid for the time spent at the
place of employment. Straight time compensation is common for part-time workers and other
clerical personnel important in back-office operations of financial institutions. Most hourly
workers are not highly paid and workers and employers share the benefits of flexibility in staffing
size and hours worked. Straight time workers rarely develop a sense of commitment to their
employers and often quit as their skills reach desirable levels, especially in the operation of
equipment like check-encoding (proof) machines.
Salaried workers earn a set amount of money over a fixed time period, say a week or a
year. They are expected to show up on time and do pretty much as told in the organization.
Straight salaried workers have a larger fixed commitment from their employers in terms of length
of time commitment and more to lose in terms of future income committed by the employer by
While the effort levels demanded of salaried workers fluctuate with demand for their
services, the fixed commitment from the employer has the benefit of reducing uncertainty about
future income. Most salaried jobs involve periodic (often annual) evaluation and reviews by
management. Salaried jobs offer workers some incentive to improve skills to order to maximize
their chances of retention and salary increases.
Piece work is the simplest pay scheme where workers are paid per unit of output. A few
financial institutions use piece work compensation to pay workers for simple jobs like opening
new accounts or completing money changing transactions. The problem with piece work is that
if quality is a factor, workers will sacrifice quality for quantity of output. For example, if tellers
are not responsible for cash errors, piece work would encourage sloppy cash controls.
Piece work compensation schemes are probably underutilized in the more routine and less
quality sensitive areas of financial service firms, like back offices. In many cases financial
service firms are reluctant to offer such payment schemes because of uncertainty about costs or
revenues with output upon which piece work schemes could be based. We discussed cost and
output measurement problems in the previous chapter. Dangers of production based incentive
schemes is that workers will go to extremes to earn money producing unprofitable goods or
services. The need to optimize employee performance is yet another motivation for studying
costs and measuring output better as discussed in the previous chapter.
In sales activities, like loan agents obtaining mortgage loan applications, insurance agents
selling policies, or securities brokers selling stocks and bonds, workers tend to minimize work
effort unless compensation is related to production. Simple output sharing compensation
arrangements discussed above are reflected in worker pay in selling of financial services like
mortgage loans, insurance policies, or trading activity. Many mortgage lenders pay sales agents
commissions based on loan production. Insurance agents often keep a major portion of the first
year's insurance premium. Most securities firms compensate brokers by splitting commissions
on trades. In many cases, these employees are guaranteed a minimum salary or "draw" over
some time period. These arrangements are similar to the tenant farmer sharing risk with the
landlord because of the minimum pay and commissions related to loan volume, insurance sales,
or stock trading.
When commissions or other pay schemes similar to piece work are used, most evidence
suggests that workers respond energetically. This may cause financial firms problems. Insurance
agents are accused of being "replacement artists" causing insurance customers to buy new
policies unnecessarily replacing economically valuable existing policies. Mortgage sale agents
are accused of helping unqualified borrowers obtain mortgages. Brokers are accused of
generating excessive trading activity by "churning" their customers' accounts to inflate
commission income, potentially harming their employer's reputation and making them liable to
lawsuits. This is a danger of piece work compensation when quality of service matters to the
employer. More sophisticated compensation arrangements might improve on these pay methods,
as discussed below.
Participation contracts make employee pay a function of some unit or division profits or
other performance measures. For example, employees of securities trading firms are often
compensated based on the profits of their specialized unit, such as a Treasury trading department.
Managers of branches of banks and other financial service firms are often paid on the basis of
their branch "profits." Paying workers based on performance aligns their interest with the
performance of the division of the firm where they work. This can produce employee
performance in line with the firm's goals if maximum profit center earnings are in line with
maximization of the firm's overall value.
Participation pay based on unit profits introduces problems in calculations of profit center
earnings, a point we discuss below in conjunction with transfer pricing. Profit participation in
units where profitability is hard to assess because of difficulty in setting transfer prices or poor
control systems can lead to unchecked employee performance harming the firm. For example,
Joseph Jett was a Kidder Peabody bond trader who falsely inflated trading profits $350 million
(while covering losses of $100 million) to obtain multimillion pay in 19923.
Bonus systems are usually combined with low salaries and compensate employees on the
basis of exceeding some target levels of production or performance. For example, a life
insurance district sales manager might be paid a $5,000 bonus for every ten percent increase in
policy sales over the preceding year's sales. Bonus systems can be thought of as modified piece
work pay with bonuses determined by bracketed amount of production. Bonus systems are
common with managers responsible for groups or teams of workers and provide managers
incentives to urge increased productivity of subordinates.
Targets or quotas used in paying bonuses are often negotiated by the employee and the
manager. While information asymmetries may make target setting subject to principal-agent
problems, well designed bonus systems may develop information concerning expected future
activity levels useful in organization planning efforts. Bonus systems based purely on volume,
like piece work pay, can distort the quality of products sold.
A "tournament" is jargon describing employee compensation and job responsibilities
based on a contest with other workers. A worker's performance evaluated as superior to other
workers wins the promotion or job assignment or large pay increase. The evaluation scheme may
be clearly based on quantitative objectives, like producing the highest profits or sales, or it may
be based on less well defined qualitative criteria, for example demonstrating superior
management skills by turning around a failing division rather than simply managing an
untroubled division acceptably. Tournaments have the benefit of inducing competition and high
levels of effort if the prize is big enough. Losing the tournament may reduce morale for other
workers who may leave or shirk even though they performed nearly as well as the winner.
Finally, many executives receive pay in the form of stock or stock options. Stock
ownership aligns top management's interest with those of shareholders and the same is true for
stock options at conversion prices reasonably expected to be surpassed by market values under
Michael Siconolfi, "Kidder Trade in Bond Options Dismissed,"
Wall Street Journal, July 12, 1994, p. A3.
good management. The highest paid executive in Fortune executive compensation survey of 200
companies in 1993 was Sanford Weill of Travelers, Incorporated, the insurance and brokerage
(Smith Barney Shearson) firm. His total pay was $45.6 million, of which $41.4 million was in
the form of long-term incentives and stock grants. Other bankers and financial service
executives in survey also received generous stock option plans, often valued at more than
Typical compensation schemes described above can be used in combinations to solve
some principal-agent problems between financial firm employers and their workers. Other
problems may require more complicated or non-standard compensation schemes. Economists
have studied a wide range of problems involving information asymmetries and imperfect
monitoring of workers' activity and found that solutions can involve complex compensation
contracts based on a number of indicators in addition to production or time, for example the
average performance of other workers or production of similar products.
Some principal-agent problems can be resolved by workers choosing among a menu of
compensation agreements which induce them to reveal information about themselves or signal
assessments of important market information or information concerning production conditions.
Offering workers a choice of compensation schemes is rare in financial service firms. Financial
service firms in general have not been sophisticated in developing compensation schemes.
Several dimensions of financial services discussed in Part II of this book differentiate
production of financial services from manufacturing or other service production. The many
activities in the value chain required to produce financial services mean the value produced by
employees may not be effectively be measured by a single output measure. The difference
between financial services and other production or sales environments is undoubtedly relevant to
the design of optimal compensation schemes for workers in the financial services industry.
Optimal compensation of financial service employees may not be easily captured by a relation to
a single measure, for example the number of loans, insurance policies, or trades sold per period.
Analyses of compensation in production contexts focus on production over short time
intervals, like the growing season in our simple tenant farmer example. Financial services like
credit services or asset management services take place over long periods of time. These
financial services are part of a long-term relationship possibly involving a number of products or
services not fixed at the outset of the relationship. Financial firms' reputations are often
important to their success in attracting new business.
Because of long-term committments, variable level of demand for services by customers,
and the importance of reputation, officials of financial service firms must sustain a level of
customer satisfaction and trust sufficient to optimize the value of customer relationships. Piece
work or commission based pay based on transactions can result in workers overlooking the
benefits of long-term relations where future business may be infrequent but profitable. Many
firms consider their employees' relations to their customers to be among their most important
assets, a source of value in view of future business from financial service demands in the future
and from referrals.
The time horizon over which financial services take place also means that future
performance under financial service agreements impacts the value of the relationship.
Employees must monitor customers to assure that the relation works as expected, for example in
credit arrangements or insurance contracts. One way to induce employee engagement in long
term performance of customer relationships is to have workers take a quasi-equity position in the
profits from customers, as in a portfolio of customer loans. If the worker's pay depends not only
on the making of the loan but the future performance of the loan, as it would if the worker
"owned" a share of the profits from customers, incentives of employees and financial firms could
be more effectively aligned.
Another complexity of financial relationships is that there is often more than one service
or product. Bank calling officers are concerned primarily with their business customers' deposits
and credit needs, but other products, like trust services and securities trading services, are also
available. Compensation schemes must be tailored to induce financial service firm
representatives to sell the whole range of products available or to induce referrals to other parts
of the financial firm organization. For example, home mortgage sales people paid on a
commission will neglect selling automatic deposit deductions or other services unless explicitly
motivated to do so. Cross-selling may take time and effort by workers to communicate complex
information remote from their main sales activity. Unless rewarded, these employees find their
time used more profitably to sell familiar products.
Sometimes financial service products are so different that the same compensation scheme
for sales cannot be applied to them. For example, brokers expect immediate income from
securities trades of their customers and brokers will make as many trades as they can, praying for
period of heavy trading volume. Insurance salesman take a long time to explain and sell
insurance policies to their customers, usually receiving a large commission upon sale of the
policy, often the first year's premium. For life insurance agents, a few sales can be a year's
income. The differing sales problems for brokerage and insurance means optimal compensation
schemes and sales culture must differ. The difference in insurance and securities sales is one
explanation offered for Merrill Lynch's lack of success in selling insurance through brokerage
Many financial services require a team effort. Motivating workers to function effectively
in teams may require compensation contracts which include not only an individual worker's
performance but also assessment by team members or a participation contract based on team
performance. For example, investment banking fees generated by a team presentation to a
corporate client could the basis for individuals' pay. The total compensation could be linked to
the number of relevant investment banking services sold so that teams would be motivated to sell
a wider range of products that just those of interest to the team members.
Finally, employees of financial service firms must be creative and innovate. New
financial products and services is where the highest profitability will be found in the future. Pay
schemes which encourage innovation and risk taking in product development or new customer
services are essential in the financial services industry of the future. Compensation arrangements
for employees expected to innovate must tolerate unsuccessful efforts at innovation if a large
number of ideas is required to identify valuable new products. Financial service firm workers
must be willing to take the risks of experimentation and failed ideas at the same time as they are
motivated to innovate. For example, employees could receive time off to develop new ideas with
pay and be given participation in new product profits or other inducements to come up with new
ideas. Compensation arrangements like this are rare in financial service firms today.
24.2 Financial Service Firm Organizations
Human resources are a critical input into the provision of financial services. How can
these resources best be organized in order to motivate workers to provide these services
effectively and responsibly? Given the complexity of many tasks in this industry, how can
workers' performance be monitored? The legal form of business organization--corporation,
partnership, sole proprietorship, or mutual--has a role in assigning responsibility and allocating
rewards in complex organizations of individuals. Grouping of workers into divisions or other
firm units with reporting relationship and performance responsibilities influences how efficiently
workers interact to achieve the firms' objectives. We discuss the legal structure and forms of
organization in turn.
Legal Form of Financial Firms
Regulation and taxation explain many legal aspects of organizational forma and
complexity of financial service firms. For example, until 1994 banks in the United States had to
have a separately incorporated bank subsidiary in every state where it wished to take deposits.
Insurance companies often have many subsidiaries to do business in different states. Credit
unions must be mutually owned. Holding company structures are widely used to evade
restrictions on activities of regulated financial service firms, like banks, thrifts, and insurance
companies. Legal and regulatory considerations explain some, but not all, of the legal structure
of financial service firms.
Business firms of all types can be considered as a system of agreements and contracts
binding a variety of parties together in producing goods or services -- suppliers, workers,
managers, customers, creditors and investors. All these parties will have differing interests and
desires. Each party works naturally enough to maximize his or her own welfare, leading to
conflicts of interest. Resolution of these conflicts are studied using principal-agent analysis as in
the previous section.
A solution to conflicts affecting the efficiency and reliability of an organization may
depend which is the best legal structure for the tasks performed by the organization. According
the analysis of Fama and Jensen (1983a, 1983b) presented in Chapter 3, the corporate form of
organization developed historically to undertake large, complex investments, spread risks, and
limit liability to initial investments. Their argument is that corporate structure is favored by
corporations like commercial banks requiring skilled professional management to work with
complex nonmarketable assets like commercial loans. Owners of corporations claim income left
over after paying all other parties with claims on the firm, like workers and creditors. Owners of
corporations elect a board of directors with the power to fire management to oversee their
Fama and Jensen note that the partnership form of organization requires that partners
share liability and income with professional colleagues, in contrast to corporations. Fama and
Jensen argue that partnerships can be effective organizations when firms provide professional
services like investment banking. Shared liability bonds partners to the highest level of
performance because each partner's performance places the firm and its reputation and future
business at risk. Investment banks and brokers are frequently organized as partnerships.
In the Fama-Jensen argument, mutual organizations are efficient when assets are simply
evaluated or readily marketable. Mutual organizations can be efficient ways of organizing
activity when agency problems are not caused by asset or service complexity. Mutual fund
shares sold to investors and deposit-taking institutions organized as mutuals, like thrifts and
credit unions, survive because their assets are marketable or relatively simple and mutual
shareholders or members can enforce management discipline by threatening to withdraw funds.
Other arguments have been made for the mutual form of organization relative to stock
companies. Rasmussen (1988) uses an agency argument to show that investors may trust mutual
organizations to invest in safer assets than stock companies. The reason for the trust is that the
mutual company managers will lose their jobs if the firm goes bankrupt from risky investments
they have made. Since there are no shareholders in a mutual who could benefit from risky
investments with potentially large payoffs (and only limited losses due to limited liability),
managers have no incentive to take risks. Mutuals will be safer places to place savings and
hence are an efficient form of savings for risk averse individuals.
Another explanation for the existence of mutuals, made in this case in the context of the
insurance industry, has been advanced by Mayers and Smith (1986). These authors argue that in
insurance it may be more costly to resolve the conflicts between creditors and owners than
between customers and managers of firms. They argue that it may be more efficient to have only
customer as owners, that is, a mutual form of organization. If policyholders are simultaneously
owners of the corporation, the need for costly monitoring of stockholders to prevent
expropriations of wealth from customer/creditors is eliminated. Customer owners may
concentrate on overcoming conflicts between customer needs and management interests.
Another argument explaining the existence of mutual firms, more popular than academic,
is that these organizations are run for the benefit of management. According to this view,
managers take control of mutual firms and pay themselves large salaries and provide themselves
with desirable perquisites. For example, the Wall Street Journal reported on large pay raises at
mutual insurance companies in 1982 as follows:
The widespread increases, disclosed in reports filed with state insurance
regulators, add to the questions about who really controls mutual insurers. The
companies are owned by their millions of policyholders rather than by
stockholders. But critics say they are often run for the benefit of entrenched
managements, which leads to inefficiency and cronyism. [May 14, 1982, p. 27]
Rasmussen, cited above, provides evidence of the low participation of mutual deposit-taking
institution depositors at annual meetings, and suggests that management really does control most
An argument that mutual firms exist because of management greed and demands for
perquisites does not explain why mutuals develop in the first place. Management exploitation
arguments do not explain how inefficient high-cost mutuals survive competition from corporate
organizations in the same business which can underprice and defeat inefficient competitors in the
market for financial services.
Other arguments for existence of mutual organizations in financial services are first in
many observers' minds -- the tax treatment of mutuals. Many mutuals are tax exempt or tax
advantaged. Credit unions do not pay taxes. Mutual funds do not pay taxes if they pass their
income through to owners. Other mutuals are taxed, but usually income is only taxed to the
extent that the income is attributable to owners. The partnership form is also tax advantaged in
that partners pay taxes as individuals only once on their share of the partnerships income.
Corporations pay corporate income taxes and owners pay taxes on salaries from the corporation
Tax rules discussed Chapter 17 make mutual versus corporate or partnership organization
important economically in terms of tax treatment. Nonetheless, economic arguments based on
agency and management issues is probably more important in explaining the long history of the
joint existence of many forms of organizations providing financial services. Mutual and
corporate forms have coexisted in periods predating the corporate and personal income taxes and
have survived innumerable changes in the tax code.
A wide variety of organizational structures are observed among financial service firms.
Sizable organizations must be broken down into coherent units. These units may be the basis of
profit centers used in motivating workers and monitoring performance. Organizational structure
can be a means of lowering costs through improved coordination and communciation or focusing
effort and increasing revenues as part of an firm's strategy. The underlying idea of designing
organizations is coordinating the efforts of many workers to foster efficiency and smooth flows
of information and resources within the organization to accomplish goals of the firm.
An important aspect of an organization is the degree of centralization desired by
managers or investors. Centralized organizations use organizational structure to control and
monitor implementation of policies. Decision making in centralized organizations tends to be
top down, with business initiatives generated by top management and performance at lower
levels reviewed and approved in terms of organization objectives.
Decentralized organization structures encourage initiatives at lower levels of the
organization. Decentralized management emphasizes coordination rather than control of
individual units. Decentralized structure fosters flexible responses to local situations.
Performance of decentralized units is in terms of broad organizational targets rather than
implementation of specified business plans developed by the firm.
The tradeoff between tightly centralized organizations which monitor and control
performance versus decentralized business units with the ability to respond quickly to local
business opportunities will be the crux of financial institutions organizational development in the
future. Changes in communications and computing technology and competition are changing
dramatically the economics of coordinating, monitoring, reporting and controlling diverse
business units in both centralized and decentralized systems. Centralized and decentralized
organizations may be best under differing competitive and technological circumstances.
Examples of centralized and decentralized organizations can currently be observed in two
United States banking corporations. NationsBank, an outgrowth of North Carolina National
Bank (NCNB) is highly centralized under its chairman High L. McColl, Jr., and controls its
subordinate divisions tightly. A merger of a banking unit into NationsBank is followed by
installation of managers from headquarters and standardization of systems. BancOne of Ohio, on
the other hand, is highly decentralized under its chairman John B. McCoy. BancOne keeps local
management in place following mergers and encourages local initiatives. Both organizations
install financial control systems after mergers but NationsBank integrates subordinate units into a
tightly controlled system while BankOne standardizes reporting to enable local managers to
compare and evaluate their performance.
Three ways of organizing business organization are commonly observed in financial
services: (1) customer or regional groupings; (2) product divisions; and (3) functional areas.
The value creation matrix of Chapter 2, repeated here as Table 24-2, can be useful in
understanding these organizing principles for workers and resources within either centralized or
decentralized financial institutions. Product divisions in financial service can be visualized in
Table 24-2 as grouping employees participating vertically into business units, for example
transaction based services. Functional organization corresponds to organization or workers
horizontally in the table according to the activity they perform in the value chain, for example
grouping employees with responsibility for communication and marketing in several product
lines into business units. Most organizations use combinations of the three organizational
principles to group resources and workers.
Customer groups are used by many financial firms to organize activities. The goal is to
be more effective in dealing with market segments, for example retail customers, by grouping
experienced and trained personnel with similar market focus together. Many firms rely on
geographical breakdowns of organizations to maximize efficiency in gathering local area market
information and communicating effectively with financial services customers with similar needs.
Financial institution branches in different geographical regions can be focused on particular
market segments, like retiree household depositors in Florida and mortgage borrowers in Utah
and Arizona or middle market commercial borrowers in the Silicon Valley. Commercial banks
and thrifts often use branches or groups of branches as business units. Many financial
institutions have representative offices in large cities or different regions.
Product divisions are another common organizational form. For example, employees
responsible for specific casualty insurance lines may be grouped together in a unit of a diversified
insurance company. Trust services and corporate lending, representing asset management and
wholesale credit services on Table 24-2, are organizational units in many banks. The advantage
of product organization is improved communication and shared experience of workers
specialized in performing all activities in the value chain for similar products. The disadvantage
of product organization is a narrow focus possibly insulating employees from developments in
other markets. For example, an organization focused on deposit taking and transaction
processing may not see asset management services like mutual funds as a possible substitute and
competitive threat. Product organization can also obscure potential synergies from joint
production of activities in the value chain for several financial services, for example by cross
selling products as part of marketing and communication with customers.
Organizations can also be grouped by functions. Functions used to organize employees
might be finance or marketing areas, funding and investing and communications and marketing
in terms of the rows of Table 24-2. An advantage of functional organization is to pool
employees with special expertise and experience, like marketing research. Functional
organization may focus effort effectively but runs the danger of separating activities which are
produced jointly in offering financial services, like production and delivery and marketing and
communication of retail services through branches. Functional organization often works best in
centralized organizations with a narrow product line where controls between organizational units
can be imposed from above without obscuring important product or market relationships.
In setting up or changing organizational structure, the major focus of financial service
firms should be to maximize the productivity of employees. The organization must reflect the
most important information flows and cost synergies from shared efforts in performing the
activities in the value chain necessary to offer financial products and services. The structure
should induce desired levels of communication and team work, whether in servicing customers,
developing new products, or in implementing new procedures. Factors like team identification
and esprit de corp are important considerations in organizing activity.
24.3 Transfer Pricing
Multidivisional financial organizations are linked by a shared interest in many resources
and personnel working in related areas. For example, commercial banks and thrifts lend money
raised from depositors. If the depositors are household customers serviced by a retail banking
division and lending is to the wholesale market in the corporate lending division, resources must
flow between organizational entities. The prices or rates which are used to account for resource
transfers within organizations are called transfer prices. These prices may be important in
determining the reported profitability of profit centers in different divisions of complex
The importance of transfer pricing can be demonstrated with a simple example based on
real problems facing many banks and thrifts. There are two profit centers in the example, retail
deposits and corporate lending. Balance sheets and income statements for the two divisions are
presented in Table 24-2. The retail division has more deposits than it can lend to borrowers so it
makes funds available to the corporate division which can lend more money than it can raise in
deposits4. The interest rate on loans is 6 percent. In the example, the transfer price on the funds
between divisions is 5 percent. Clearly the profitability of the two divisions depends on the
transfer price. At five percent, the retail division is more profitable. At a transfer price of four
percent, the corporate division would be more profitable.
The importance of transfer prices is based on several factors. These factors are: (1) the
effect of transfer prices on the allocation of resources in the firm and maximization of value for
the entire firm; (2) the effect of transfer prices on the incentives and compensation of managers
based on divison profits; and (3) the production of information concerning the relative efficiency
of divisions and internal sources of the organization's competitive advantage relative to market
The interest rate or service cost of deposits is assumed to
be three percent for convenience.
competitors. We discuss each in turn.
Firms are formed to reduce transaction costs of doing business5. More efficient
information flows and efficiency from better coordination of activities and economies of scope
can be realized within organizations reducing the costs of transacting business having different
components or stages in production. In large, multidivisional organizations, specialization by
customers, products, or function discussed in the previous section results in resource transfers
If transfer prices are not correct, division managers may overproduce products or services
using undervalued inputs. They may produce the wrong products by responding to transfer
prices which are too high, increasing profit center performance but reducing the firm's profits
overall. If transfer prices for activities produced within the firm are set too high, profits may be
lower than those possible using external suppliers in producing products or services. If transfer
prices are set too low, the firm may not realize maximum profits from demand or technological
dependencies linking production of products or services.
Most current of the current discussion of transfer pricing currently in the United States
concerns the role of transfer prices in shifting income tax liabilities across international borders.
Transfer prices for goods shipped internationally can be used to shift profits of multinational
corporations. In the United States, section 482 regulations of the Internal Revenue Service (IRS)
outlines acceptable methods for transfer prices used in filing company taxes when resources are
shipped across international tax boundaries. In our above example, imagine that retail deposits
are from an United States subsidiary and loans are in a low-income tax foreign country subsidiary
of a multinational bank. Lowering the credit on deposits would move profits to the foreign
IRS rules provide methods to benchmark acceptable transfer prices. These methods are
difficult to apply and lead to a great deal of conflict and litigation. In 1987, for example, Nissan
and Toyota automobile companies from Japan paid the IRS $604 million in taxes on income
reduced by high transfer prices on autos shipped to the United States6.
The focus in this chapter is on the economic incentive effects of transfer prices although
many of the issues are the same as in international tax disputes. The example of the deposit-
taking firm reveals the importance of transfer pricing to the assignment of taxable profits earned
in multidivisional corporations. When management compensation or performance evaluation is
based on divisional profits, managers' pay or promotions in some divisions will be increased and
others decreased by changes in transfer prices, just as tax liabilities are changed across
international borders with transfer price adjustments. In many financial institutions, the fixing of
transfer prices like the cost of funds in our example is a serious political problem because of
repercussions on management compensation.
Hirshleifer (1956) presents an economic analysis of the effects of transfer pricing within
These ideas stem from Coase (1937) and elaborated by
See Granfield and Weston (1990) for an review and analysis
of IRS methods for reviewing transfer pricing. The Japanese tax
agency refunded the $604 million to the auto companies.
multidivisional firms where there are no important management incentive effects. The emphasis
of Hirshleifer's analysis is multidivisional firm managers who maximize profits of their divisions
and make decisions based on transfer prices for intermediate goods. Our discussion assumes that
a central management sets transfer prices to maximize profits of a firm facing external markets
and internal production conditions. Hirschleifer stresses that either final or intermediate good
division managers motivated to maximize firm (rather than divisional) profits would make the
same transfer price decisions. Table 24-4 summarizes the results of Hirshleifer's analysis we
discuss in the following.
According to Hirshleifer's analysis, intermediate good transfer prices in general should be
set equal to marginal costs of the intermediate good at the firm profit maximizing production
level. Considering retail deposits to be an intermediate good from our example, transfer prices
should be set at the marginal cost of funds which maximizes the deposit-taking firm's total
profits. The third column of Table 24-4 provides interpretations of Hirschleifer's analysis in
terms of our deposit-taking financial firm example. In case (1), loan deposit ratios are fixed and
in case (3) the market for deposits is upward sloping meaning the deposit-taking firm has market
power in its deposit market. In both these cases, the optimal transfer price for the deposit
division is where the marginal cost of deposits equals the marginal revenue from loans. In case
(2), the deposit market is competitive and the optimal transfer price is the market rate on
deposits. Since interest rates are fixed by competitive conditions in case (2), marginal deposit
costs for the firm are equal to market rates. Therefore for cases (1) through (3), marginal cost
transfer prices are optimal.
In cases where there is dependence between the intermediate good and final good
demand, for example when deposit taking is not independent of loan demand, transfer pricing is
more complicated. This is case (4) in Table 24-4. The effect of the deposit division's actions on
the profitability of lending must be considered in the setting of transfer prices. If deposit-taking
tends to increase the demand for loans, perhaps by expanding customer contacts, optimal transfer
prices will be between market rates and the marginal cost of funds. In the deposit-taking firm
example, the credit rate on deposits should be above the marginal cost of funds to induce
managers to increase the supply of deposits to levels maximizing joint profits from deposits and
Hirschleifer's analysis stresses the importance of using marginal or incremental costs in
setting tranfer prices under competitive conditions where intermediate and final products are
independent technologically and have no market dependencies, cases (1) to (3) in the table.
Hirshleifer raises an important point concerning long-run effects of transfer prices when fixed
costs are considered in addition to variable costs. Divisional profits including allocations for
fixed costs may be negative using transfer prices based on marginal costs which maximize total
firm profits. It cannot be determined whether divisions reporting losses when fixed costs are
included should be closed without looking at the total firm profit picture including fixed costs for
other divisions. Losses from one division may subsidize larger profits in other divisions. This
could meant that divisions reporting losses when fixed costs are counted contribute to overall
firm profitability in the long-run. Optimal transfer pricing may not produce division income
statements useful in identifying firm wide competitive advantages.
In cases (1) to (3) in Table 24-4 analysed by Hirshleifer, marginal cost transfer pricing is
a robust approach for optimizing short-run decisions (ignoring costs) with transfer prices.
Technological and demand interdependencies complicate analysis of optimal transfer pricing.
Hirschleifer's analysis also ignores incentive effects on divisional managers who must expend
costly effort to produce in response to transfer pricing signals from central decision makers in the
organization. Marginal cost or market pricing may not be optimal when accounting for these
effects, as in case (4) of Table 24-4.
In an economic analysis of divisional managers responding to transfer pricing in
producing intermediate inputs to a final output for a firm, Besanko and Sibley (1991) show why
division profits should be calculated using transfer prices departing from market prices or
marginal costs. Their analysis focuses on the principal-agent problem caused by managers
compensated by profit-sharing within the firm who are buying inputs from the external market
and producing outputs for the firm. The analysis is interesting because it shows the effects of
technological dependencies between intermediate and final goods and management incentives on
optimal transfer prices. Differences in transfer pricing from market prices are necessary to make
the incentives of the manager responsible for producing the intermediate input align with those of
the organization as a whole.
The Besanko and Sibley analysis demonstrates how transfer prices induce managers of
divisions to expend optimal effort while accounting for technological interdependencies between
outputs for various divisions. For example, in our deposit-taking firm example, assume that
managers of deposit-taking divisions stimulate loan demand for the lending division in the
process of gathering deposits. By providing good service and calling on deposit customers, loan
demand from the institution is increased. This causes a technological or market dependence
between deposit and lending divisions. In this case, the deposit-taking division should receive
higher transfer rate than market rates or marginal costs to reflect the positive effect of deposit
services on loan profitability. By paying higher rates on deposits, managers in the deposit
division have greater incentives to increase their efforts. This is similar to case (4) in Table 24-4.
Transfer pricing is clearly important in all financial services. We can consider
interdivisional transfers between deposit services and credit services as in our example as transfer
prices between columns of the value creation matrix in Table 24-2. Additional examples are
transfer prices on funds provided by insurance reserves used by credit service or asset
management service divisions of insurance companies and transfer prices assigned to transaction
processing costs associated with securities trading services offered by brokerage firms. Many
other financial service firm examples of transfers between columns of Table 24-2 are possible.
The importance of transfer pricing in financial services will continue to increase as
financial services respond to competitive and technological developments. If we view separating
the performance of activities listed in the rows of Table 24-1 by different firms or divisions as
unbundling financial services, costs and transfer prices between rows of the table become
important in determining competitive advantages in value chain. For example, decisions to
produce internally or outsource activities require knowledge of optimal transfer prices for
activities produced within a firm and market prices for activities produces by external suppliers
and reasons for departures of internal transfer prices from marginal costs, for example demand or
Compensation of workers efficiently providing activities in the value chain necessary to
offer financial services will depend on profit centers using optimal transfer prices. Synergies or
technological and demand interdependencies can be the basis of value creating strategies if they
can be identified, if costs measured adequately, and if workers can be motivated to perform as
required. Use of optimal or at least good transfer prices are central to accomplishment of value
creation in strategies based on transfers or resources between the rows or columns of the value
Transfer Prices in Practice
How are transfer prices set in multidivisonal organizations? There are three practical
ways to arrive at prices or rates between divisions of large organizations7: (1) market
benchmarks; (2) political negotiations; and (3) cost estimates. We discuss the advantages and
disadvantages of each of these methods in financial service firms. These three methods may be
compared to criteria applied by the IRS to transfer prices used by multinational firms. The IRS
guidelines are attempts to determine whether transfer prices used in international transactions are
legitimate results of one of the three methods listed or are set arbitrarily to minimize global tax
liabilities. In assessing transfer prices reported on tax returns, the IRS also looks at return on
asset and return on equity calculations and market prices and markups for other firms. The IRS
is not concerned about optimal or efficient organization of activity.
The advantage of market based transfer prices is that they represent objective measures of
alternatives to internal production of intermediate resources. Even if market measures are not
used in transfer pricing problems, managers at all levels of organizations should be aware of the
prices of substitutes available from the market. Knowledge of market alternatives can prevent
negotiated prices from distorting allocation of resources too much and provide a useful
benchmark for cost based estimates. Managers can use market prices in negotiations effectively
by simply using the argument that they could obtain resources, like funds or computer power,
from outside vendors at quoted values.
Market benchmarks have two disadvantages. First, in line with the discussion above,
market based transfer prices may not induce optimal effort and production in multidivisional
organizations. Second, many transfers between divisions do not have perfect substitutes in the
market. Many firms adjust transfer prices to account for market phenomena, for example by
adjusting internal deposit transfers for differences in maturity using adjustments from the term
structure of interest rates. Comparisons of internal transfers and adjustments to market
quotations may not adequately capture the economics of internal resources. For example, retail
transaction deposits may be cheap because core deposits have low turnover despite their
availability on demand.
Negotiated transfer prices have the advantage of accounting for special situations in the
organization. Negotiations can in fact replicate internal markets where the unique economic
tradeoffs of the organization are explicitly reflected in transfer prices. In many instances, for
example in pricing computer services or technical advice, transfer prices may effectively signal
See for example Horngren and Foster (1991), Chapter 27.
Granfield and Weston (1990) detail IRS acceptable methods.
development of corporate strategy, say by computerizing diverse operations. To the extent that
internal users are competing for divisional resources and transfer prices reflect internal demand,
management can determine the efficiency of relying on internal units to perform activities instead
of using market substitutes available externally.
The biggest problem with negotiated transfer prices is that they are subject to political
manipulation. Most retail bankers complain that their operating units are not given sufficient
credit for resources because of transfer prices fixed to benefit more powerful and entrenced
management in areas like corporate lending or investment banking services. Political transfer
prices can blunt management incentives to provide intermediate resources efficiently.
Cost based transfer prices are subject to all the problems of cost measurement discussed
in the previous chapter. The rule for cost based transfer prices is8:
Transfer Price = Variable Cost + Opportunity Cost 24-1
In terms of our cost measurement discussion, both elements of this equation are difficult. The
rule includes as a first element variable costs. While variable costs avoid problems of overhead
allocations included in average cost methods, variable costs are still difficult to measure as
discussed in the previous chapter. The opportunity cost component of transfer prices is the most
difficult problem. Opportunity costs are defined as what the organization as a whole gives up or
gains by using internal resources. This component will include the incentive and demand and
technological interdependencies discussed above. Estimates of this component will be very
difficult for financial service firms but are essential in setting optimal transfer prices.
Transfer pricing is at the core of efficient operation of multidivisional financial firms.
None of the three methods of establishing transfer prices can be relied on to produce efficient
allocation of effort in all instances. Pragmatic managers will be aware of the effect of transfer
prices on decision making and compensation and derive estimates from all three methods.
Transfer prices will be reviewed and evaluated constantly as changing market environment and
competition affect the competitive advantage of financial firms.
1. Growth and Change
This book began by discussing the many changes taking place in financial services. It
must end by emphasizing the importance of flexibility and adaptability of firms and their
managers and employees in the dynamic environment of the future. Financial firms must be
organized to change. Change can mean growth, shrinkage, broadening or narrowing focus. The
importance of organization is that information concerning competitive advantages or ephemeral
market opportunities must be communicated and acted upon quickly. Experimentation, failure,
and false starts will all be necessary to find the most effective strategies.
How can firms organize effectively to accomplish the agility necessary to respond to
rapid market developments? There is no single answer. Decentralized organizations can be
responsive to local market developments but foster rogue behavior and inefficiences from poor
See Horngren and Foster (1991), p. 863.
coordination or monitoring. Centralized organizations can be unresponsive and slow to process
developments into market responses. There is no single best answer. Intelligent managers must
be constantly weigh these tradeoffs as they organize effort.
Three basic characteristics of successful organizations appear to be important in the
future: (1) employee incentives and compensation based on developments they can control in
their markets or product areas; (2) good systems review identifying poor or disastrous
performance relative to winning efforts; and (3) an atmosphere encouraging experimentation and
risk-taking. New products must be encouraged both in operating divisions and research
divisions. Communication between parts of the organization must be free and candid.
Most observers of the financial services industry believe that the future will be
characterized by the consolidation of many small financial firms. These consolidations are
economically justified by synergies gained from larger organizations. Many financial institutions
identify strategic planning with acquisitions. Acquisitions within markets and across
international boundaries will be an even more common phenomenon in the future.
The underlying issue is growth. The presumption is that growth is good despite this
book's emphasis on return on investment. Growth is only good when it increases return on
investments or maintains excessive returns. Growth can be realized internally and externally. In
any case, growth should be built on a strategy of building on strengths and shedding areas of
weakness where the criterion of performance is return on owners' investments.
Mergers and acquisitions have been common in financial services in the last decades. We
began the book with some of the more notorious examples. Mergers and Acquisitions magazine
has ranked commercial banking in the top three industries in the United States by number of
mergers in each of the last ten years.
What has been the result of these mergers? As discussed in the final section of the previous
chapter, there is little evidence to support the argument that investors gain by mergers producing
larger banking institutions. Problems from integrating management and workers into new
organizations such that incentives and skills of merged firm employees maximize efficiency of
the organization are enormous.
We conclude our analysis of the financial services industry with a challenge to future
managers. Survival and success in the future can only be based on management decisions
creating value through increased revenues or reduced costs. All decisions by good managers
must identify and justify gains from consolidations. Managers proposing growth strategies must
show explicitly the synergies to be obtained by mergers or internally funded growth. Not only
must these synergies not be left as vague assurances of the benefits of size or diversification but
the synergies must be explicitly related to reasonable revenue increases or cost reductions.
Future managers are challenged to demonstrate convincingly the impact of their decisions on
return on investment. Simple arguments based on economies of scale and scope are not a
sufficient justification for internal or external growth.
This book is intended to provide future managers with a framework for developing
arguments providing economic justification for financial service strategies in the future.
Managers must be able to argue persuasively the value of growth or diversification from potential
demand or production intedependencies between and within the rows and columns of the value
matrix in Table 24-2. With labor and human capital intensive activities dominating the value
chain in financial services, with the role of information distribution and analysis in the face of the
technological developments favoring inexpensive information distribution and powerful local
information processing power, managers must show the advantages of larger scale operations
concretely. The best managers can and will be able to use the value matrix and the analytical
tools provided in this book to show the gains from growth and/or change policies they advocate
in terms of return on investment.
Value produced by financial service firms comes largely from the skills and coordinated
activities of workers. Motivating workers with incentive based pay can be very important even
though creative compensation schemexs are currently infrequent in financial services.
Organizational structures affect the ability of good people to work and communicate in firms.
Organization is important to foster a spirit of innovation and experimentation required by the
further evolution of the financial services industry. Transfer pricing within large scale
organizations is a way to communicate top management objectives, to provide incentives to
workers, and to identify competitive advantages of financial service firms. Given the importance
of pay, organizations, and performance accounting in financial firms, strategies for growth and
change must be carefully formulated in terms of synergies in the value matrix to be successful.
The most successful firms, those with the highest return on investment, will not be built on
hunches and hopes.
Besanko, David and David S. Sibley. 1991. "Compensation and Transfer Pricing
in a Principal-Agent Model," International Economic Review, Vol. 32, No. 1
(February), pp. 55-68.
Coase, Ronald. 1937. "The Nature of the Firm," Economica, Volume 4, pp. 386-
Fama, Eugene F. 1980. "Banking in the Theory of Finance", Journal of Monetary
Economics 6, pp. 39-57.
Fama, Eugene F. and Jensen, Michael C. 1983a. "Separation of Ownership and
Control", Journal of Law and Economics (June), pp. 327-349.
Fama, Eugene F. and Jensen, Michael C. 1983b. "Agency Problems and Residual
Claims", Journal of Law and Economics (June), pp. 327-349.
Granfield, Michael E. and J. Fred Weston. 1990. "Effective Section 482
Enformcement and the Japanese Auto Cases," Tax Notes, March 5, pp. 1187-
Hirshleifer, Jack. 1956. "On the Economics of Transfer Pricing," Journal of
Business, Volume 29, pp. 172-174.
Mayers, David and Smith, Clifford. 1986. "Ownership Structure and Control:
The Mutualization of Stock Life Insurance Companies", Journal of Financial
Economics 16, pp. 73-98.
Milgrom, Paul and John Roberts. 1992. Economics, Organizations, and
Management. Prentice-Hall, Englewood Cliffs, New Jersey.
Rasmussen, Eric. 1988. "Mutual Banks and Stock Banks", Journal of Law and
Economics (October), pp. 395-421.
Sappington, David E. M. 1991. "Incentives in Principal-Agent Relationships,"
Journal of Economic Perspectives, Volume, Number 2 (Spring), pp. 45-66.
Williamson, Oliver E. 1975. Markets and Hierarchies. The Free Press. New
Type of Compensation Description Where Observed
Straight time Hourly or day payments Part-time clerical workers,
when on the job many back office functions
Salary and review Compensation fixed at all Most full time jobs in
levels of output financial service firms
excepts sales and trading
Piece work, commissions Fixed payment per unit of Securities brokers,
production mortgage lenders, new
account personnel in banks
Performance participation Formula relating Trading departments of
compensation to investment banking firms
performance measure like
Bonus system Bonus based on achieving Managers of branch banks
specified targets of
Tournament Several employees compete Managers and executives in
for promotion and raise all areas of financial firms
Equity participation Stock or stock options part Top executives of most
of compensation financial service firms
Financial Services and Value Producing Activities
Services/ Credit Securities Insurance Transaction Information Advising/
Activities Services Services Services Processing Services Management
Pricing/ Loan rates Underwrit-ing
Term Setting fees
Transfer Pricing Example
RETAIL DIVISION CORPORATE DIVISION
INCOME STATEMENT INCOME STATEMENT
Interest on Loans $ 60 Interest on Loans $300
Earnings Credit (5%) $200 Interest Cost (3%) $30
Interest Cost (3%) $150 Earnings Charge (5%) $200
Profit $110 Profit $70
BALANCE SHEET BALANCE SHEET
Loans $1,000 Deposits $5,000 Loans $5,000 Deposits $1,000
Claims on Wholesale Division Liability
$4,000 to Retail
Transfer Pricing Rules from Hirshleifer (1956)
Supply and Demand Optimal Transfer Example for Deposit-
Conditions Price Rule Taking Firm
(1) Single joint level of output for Marginal cost of Determine where marginal
intermediate and final good intermediate good revenue from loans equals
marginal costs of deposits
and use transfer price for
deposit funds equal to that
(2) Intermediate and final goods Market price for Price deposits at market
produced and marketed intermediate good which prevailing rates.
independently and intermediate is equal to marginal cost
good market competitive of intermediate good
(3) Downward sloping demand Transfer price is set where Add marginal revenue from
curve for intermediate good sum of net marginal loans to marginal revenues
revenue from final good from other uses of deposits
and marginal revenue and set transfer price equal
from intermediate market to marginal cost of deposits
are equal to marginal cost at that rate.
of intermediate good
(4) Demand dependence in that Transfer price between Use of deposit to purchase
external intermediate good sales marginal cost and market securities reduces loan
reduce demand for final good price demand: set transfer price to
optimal rate for aggregate
profits somewhere between
market rates and marginal