Nof S.Y. Springer Handbook of Automation

Подождите немного. Документ загружается.

Automation in Financial Services 91.1 Overview of the Financial Service Industry 1615

A new insurance product protects a customer from

a risk they had not insured before. A new type of

loan makes funds available to borrowers with differ-

ent contract terms than existing loans. Frequently, new

products are amalgamations of other, existing products;

for example, a new banking product will often be a set

of accounts(e.g., asavings account, a checking account,

a consumer loan) with certain requirements, such as

minimum balances and minimum or maximum activity

levels. Identifying and deﬁning a new product that will

be successful, however, is extremely difﬁcult. The new

product must appeal to a customer base. It must comply

with government regulations. It must be priced effec-

tively and it will likely require new or modiﬁed systems

to support the processes for implementing it.

Because products and services are intangible, they

are easy for competitors to copy. Copying the product or

service requires knowing its terms or components (e.g.,

Table 91.1 Subset of NAICS deﬁnition of ﬁnancial services

Codes Titles No. of US Automation examples

businesses

52 Finance and insurance 471196

522110 Commercial banking 50981 Loan scoring

522120 Savings institutions 10291 Internet banking

522130 Credit unions 11 997 Analytics for customer analysis

522220 Sales ﬁnancing 1999 Automate teller machines

522291 Consumer lending 13718 Obopay (mobile payments)

522292 Real estate credit 7707

522293 International trade ﬁnancing 2163

522298 All other nondepository credit intermediation 12564

522310 Mort. and nonmort. loan brokers 49853 Online trading

523110 Investment banking and securities dealing 2267 Big Bang – automation of London

stock exchange

523120 Securities brokerage 25214 Loan origination systems

523130 Commodity contracts dealing 2339 Program and algorithmic trading

523140 Commodity contracts brokerage 720

523210 Securities and commodity exchanges 331

523920 Portfolio management 6312

523930 Investment advice 19 407

523991 Trust, ﬁduciary, and custody activities 9268

524113 Direct life insurance carriers 6408 Call center automation

524114 Direct health and medical insurance carriers 3639 Claims processing

524126 Direct property and casualty insurance carriers 5962 Workﬂow automation

524127 Direct title insurance carriers 4471 Claims fraud analytics

524128 Other direct insurance (except life, health, 2041 Risk management

and medical) carriers

524210 Insurance agencies and brokerages 56733

524291 Claims adjusting 4404

knowing the terms and structure of a new loan product

or knowing the accounts and limits for a new type of

bank account) and then implementing the processes and

systems required to support the product; for example,

Prudential offers MyTerm a new, immediate-issue term

life insurance product available through the internet to

customers of select banks and other ﬁnancial institu-

tions [91.5]. This product is available through banks. To

implement it, abank connects itssystems to Prudential’s

systems, so it will be easy for any bank to offer. For

another insurance company to offer a similar product re-

quires similar systems, but they have a model on which

to base them. It would also require an understanding of

the risk, but they would not have to start their analysis

from scratch.

In part due to the ease of product creation and dupli-

cation, at least on the retail side, entry into the ﬁnancial

services industry is easy. Starting a bank requires only

Part I 91.1

1616 Part I Home, Ofﬁce, and Enterprise Automation

US$6 million in capital. Starting a credit union requires

less. Becoming a ﬁnancial planner, in most places, does

not require formal ﬁnancial education or expertise, so

entering the market is easy and inexpensive. The easy

market entry and low startup costs increasecompetition,

since it results in a large number of competitors with

both large international players and with small local

ﬁrms.

According to NAICS [91.1], there are approxi-

mately 500000 ﬁnancial services ﬁrms. This includes

over 150000 insurance agencies and brokerages, 50000

commercial banks, 25 000 securities brokerages, and

the 7 Federal Reserve banks. These companies range in

size from Citigroup with almost US$1.5trillion is as-

sets [91.2] to the Wanda Michalenko Memorial credit

union with around US$30k in assets [91.3].

Automation plays a central role in ﬁnancial service

ﬁrms. Automation of the capture or transmission of in-

formation improves productivity. This has led banks and

credit unions to automate their core processes, stock

exchanges to automate the matching and execution of

trades, and stock brokerages to implement algorith-

mic (program) trading. These automation projects have

signiﬁcantly changed the ﬁnancial services industry.

Banks’ automation of their core systems changed the

industry cost structure and was a key determinant in in-

dustry consolidation [91.6]. Big Bang – the automation

of the London Stock Exchange trading – enabled the

stock exchange to increase trading volume from 20000

shares a day to over 550 000 shares a day [91.7], and

program trading affects the stock market’s value [91.8]

as well as accounting for over 50% of the trades on the

New York Stock Exchange (NYSE) [91.9]. Table 91.1

lists some of the types of ﬁrms that comprise the ﬁnan-

cial services industry as well as some of the ways in

which they are using automation.

91.2 Community Banks and Credit Unions

Because of the size and breadth of the industry, the

remainder of this chapter focuses on small commu-

nity banks and on credit unions. These institutions have

unique automation needs. They compete against giant

companies, such as Citigroup and Bank of America.

They need to offer comparable services at competitive

prices, but they do not have the deep pockets to fund

automation projects and innovation.

Community banks, for the purposes of this chapter,

are relatively small, regionally localized banks. There

are thousands of banks with less than US$2 billion in

assets and that serve only part of a state or regional

area. Credit unions are nonproﬁt, cooperative ﬁnan-

cial institutions with the motto: not for proﬁt, not for

charity, but for service [91.10]. Each credit union is

chartered to serve a deﬁned membership; for example,

the Houston Texas Fire Fighters Federal Credit Union

only serves ﬁre ﬁghters and their families in Houston,

TX. With such targeted markets, credit unions are small.

The largest credit unions had less thanUS$2.5billion in

assets in 2005.

Both retail banks and credit unions make their

money on fees charged for their services and on the

spread between the interest they pay and the interest

they earn. To compete, they must offer both competi-

tive rates and competitive services. Some credit unions

and small community banks offer only very limited ac-

counts, loans, and services. Others offer a full range of

accounts, loans, and services.

Banks and credit unions differ in their goals and

objectives. Banks try to minimize their cost of capital

and maximize their risk-adjusted loan rate. Conversely,

credit unions try to maximize their cost of capital

(the interest rates they pay on deposits for members)

and minimize their risk-adjusted loan rate (subject, of

course to maximizing the amount and number of loans

they make to their members). Ultimately, credit unions

are operated for their members, aligning the interests

of owners and customers, while banks are operated for

their shareholders, resulting in maximizing returns for

owners while simply satisﬁcing customer needs.

91.2.1 How a Retail Bank/Credit Union

Works

At the simplest level, banks and credit unions provide

two services. First, they provide a safe and convenient

place for people to store cash and from which to make

payments. Second, they match people with excess cash

with those who need cash and then support the lend-

ing process. Processing these sets of transactions is

complicated and frequently entails substantial automa-

tion [91.11].

Figure 91.1 shows some of the interactions and

some of the detail for processing deposits, payments,

and loans. The key points from Fig. 91.1 are that the

customer interacts with the bank in numerous differ-

ent ways through various channels, and that there is

Part I 91.2

Automation in Financial Services 91.2 Community Banks and Credit Unions 1617

Clearing

house

EFT funds

Verify funds

To customer's issuing bank

Settlement

Settlement (which also delivers funds to the vendor)

EFT

network

Customer's bank's

internet banking

service provider

ATM/POS

host provider

Pay with

credit

card

Pay with

debit

card

ACH operator

Product servicer

ACH operator

Deposit@home

system provider

Settlement instructions

for all ACH transactions

Payment transactions

Amount owed to seller's bank

Settlement

transactions

Vendor's bank

Credit card

Txns

Vendor

Transaction detail

Card, pin

amount

Card, pin &

$ amount

Deposits

cash &

checks

Payment

transactions

Credit card

settlement

Settlement

bank

Private or federal

reserve

Direct settlement

checks +

cash letter

Interbank

transaction detail

checks + cash letter

Interbank

transaction detail

checks + cash letter

Federal reserve

(National Settlement

Service)

Customer's

bank

Customer's

deposit

transaction

Deposits check

Customer

Balance

update

Checks +

batch totals

Charges/

fees

Statements

Transaction

detial

Transaction data

On us

transactions

Core

systems

processor

Acount details

Scans check

via internet

banking

Pay via

internet

banking

Pays with check

$$$ cash

$$$

Deposit cash

Application to bank

Substitute check

Item processor

Application to bank

Mortgage borker

Car loan originater

Settlement

ACH

file

All deposit

transactions

ACH

file

Card

association

network

Authorization code

Fig. 91.1 Basic bank and credit union processing

Part I 91.2

1618 Part I Home, Ofﬁce, and Enterprise Automation

a complex set of systems operated by different parties

required to support those interactions.

Customer Deposits

When a customer makes a deposit, the bank records the

deposit and updates the amount it owes the customer. To

make a deposit, a customer can:

1. Take cash to the bank

2. Deposit a check at the bank

3. Have someone (e.g., an employer) make a deposit

via automated clearing house (ACH)

4. Scan a check at their home.

Each of these requires a somewhat different process,

different systems, and potentially different companies

running the systems. When the customer takes cask to

the bank, the teller counts it and enters the amount into

their core system, which updates what the bank owes

the customer.

When the customer deposits a check, the process

is more complicated. The deposited check must be en-

coded by theteller with theamount ofthe check.Checks

are batched up and sent to item processing. Item pro-

cessing sorts the checks by paying bank and account.

This may be done by the bank or by a third party. The

bank’s core systems are updated with transaction details

and checks are processed for settlement.

Settlement can occur in numerous ways. On-us

transactions are checks written on other accounts at the

same bank. For these, the bank withdraws money from

one of its accounts and deposits it in another (assuming

there is enough money available). This is just process-

ing a set of accounting transactions. When the check

is written on a different bank, the customer’s bank can

settle with the other bank directly, through a clearing

house, or through a settlement (or correspondent) bank.

With direct settlement, the checks are presented to the

check writer or payer’s bank and that bank uses a ser-

vice, such as Fedwire, to transfer funds to the customer

bank (Fedwire is a real-time funds transfer system op-

erated by the Federal Reserve banks). With a settlement

bank or clearing house, the checks and cash letter are

sent and the clearing house nets all of the transaction

for all of the participating ﬁnancial institutions. It then

uses its own accounts or the Federal Reserve National

Settlement System (NSS) to debit or credit the correct

amounts to each participating institution. Transaction

details are also sent to each institution, so they can debit

or credit the appropriate customer accounts.

For deposits made via ACH, the payer (such as

an employer) provides its bank with instructions for

creating a ﬁle with all of the deposits it will make.

The payer’s bank – the originating depository ﬁ-

nancial institution (ODFI) – creates or veriﬁes the

ﬁle and sends it to an ACH operator. The ACH

operator routes the ﬁles to the payee (customer’s)

bank. It then settles the transactions via the Federal

Reserve.

Customers can also deposit checks from their home

by scanning both sides of the check and submitting the

ﬁles. The system supporting this process is either ad-

ministered by the Internet banking service provider or

by yet another service provider. The scanned check be-

comes a substitute check (approved by the Check 21

law) and the electronic information is processed like

a check that has made it through the item processing

stage. This information must be passed from the Inter-

net banking or deposit at-home service provider to the

bank’s system.

Note that in all of these cases, cash is not ﬂowing

from one bank to another. Only information – changes

to account balances – is ﬂowing. Also note that there

are numerous kinks in the ﬂow. The check may be

damaged, requiring manual processing at the item pro-

cessor. The check may be fraudulent or there may be

insufﬁcient funds in the payer’s account which will

prevent settlement and require the customer’s bank to

reverse the deposit. These kinks complicate the infor-

mation ﬂow and require sophisticated automation and

processes.

Payments

When a customer makes a payment, the bank records

the payment and updates the customer’s account. It also

goes through a settlement process to get the appropriate

amount to the payee. To make a payment, the customer

can:

1. Pay with cash

2. Pay with a check

3. Initiate an ACH credit transaction

4. Initiate an ACH debit transaction

5. Pay with debit card

6. Pay with credit card.

To pay with cash, a customer just hands over cash. The

only role the bank plays is providing cash to the cus-

tomer and receiving the cash from the vendor. When

paying with a check, the process is the same as when

the customer deposits the check. The only difference is

that the vendor is depositing the check. In some places,

the vendor scans the check at the time of purchase. This

converts the check to an electronic form and an ACH

Part I 91.2

Automation in Financial Services 91.3 Role of Automation in Community Banks and Credit Unions 1619

(or electronic funds transfer (EFT)) transaction occurs,

similar to a debit-card transaction.

Bank customers can also use ACH to pay bills. Typ-

ically, this is through either the Internet banking system

or a vendor’s system. When the customer uses the In-

ternet banking system’s bill-pay feature, they can set

up a recurring payment. They tell their bank to with-

draw funds and to pay a certain account. Alternatively,

they can use a vendor’s system and give the vendor

authorization to withdraw money from their account.

This frequently happens with recurring payments such

as mortgages, and insurance premiums. This process is

the same as when they deposit funds via ACH, except

they are the payer instead of the payee.

To pay with a debit card (which is similar to us-

ing an automated teller machine (AT M)), the customer

provides their card and pin (there is also a signature

version). The debit card/AT M system may be driven

by a third party. The transaction is forwarded through

the EFT network to verify that the customer’s funds are

available. If so, the account is debited and an electronic

funds transaction transfers funds to the debit card/AT M

host provider. The debit card/AT M host provider then

uses ACH to transferfunds to themerchant/AT M owner.

To pay with a credit card, the customer provides

their credit card. The transaction information is trans-

mitted through the card association’s (typically Visa

or MasterCard) network to the issuing bank’s system,

where the transaction is approved. An authorization

code is sent to the merchant. The merchant submits

batches of credit-card transactions at the end of the

day to its bank (merchant account). Settlement then oc-

curs between the merchant bank and the issuing bank

through a combination of the card association and ulti-

mately a payment network such as Fedwire.

Loans

A customer can take out a loan (the process is not

shown in Fig.91.1). There are numerous types of loans,

including mortgage loans, car loans, credit cards, and

installment loans. Taking out a loan requires ﬁlling

out an application, which can be done at the bank,

online, and at various loan brokers with whom the

bank has agreements. After the application is complete,

the customer’s credit worthiness is veriﬁed. This re-

quires obtaining information from external sources, and

frequently involves credit scoring – using the bank’s

software or acquired from an external service provider.

If the loan is approved, then it must be serviced. This is

another touch point between the customer and the bank,

requiring additional systems and potentially additional

vendors.

A ﬁnal issue is the generation of statements by the

bank. Typically this requires sorting and printing trans-

action detail and is part of the core banking system. For

some accounts that require speciﬁc balance levels and

transaction usage across bank products, at least part of

the process may be handled by an outside vendor. In

these cases, transaction detail is sent to the vendor who

determines if the product requirements were met, and if

not the appropriate fees. This information is sent back

to the bank to be included on the statements.

As is evident, there are numerous types of trans-

actions, requiring interaction among a large number

of organizations. The clearing process is similar or

the same for many transactions. This is facilitated by

standard formats for the required information, which

makes automation both easier and more effective. The

customer-facing portions of the processes are not as

standardized. This causes integration issues that are dis-

cussed later.

91.3 Role of Automation in Community Banks and Credit Unions

It is generally accepted that investments in information

technology (IT) can lead to improved ﬁrm performance

and possibly competitive advantage. Research applying

the resource-basedview of the ﬁrm hasshown that ﬁrms

with better IT capabilities perform better ﬁnancially

than similar ﬁrms with less capable IT [91.12,13]. This

is particularly important for the banking industry, which

invests heavily in IT and has increased investment from

6% of revenues in 2000 to 6.5% in 2001 [91.14]. Many

bank chief executive ofﬁcers (CEOs), however, are un-

sure of IT’s impact[91.15], while othersclaim that these

investments lead to increased services but decreased

proﬁts [91.16]. The issue is then: what drives successful

automation?

91.3.1 Deﬁnition of Success

Examining what drives successful automation in com-

munity banks and credit unions requires deﬁning suc-

cess. Depending on the purpose of the automation, dif-

ferent deﬁnitions are appropriate. Under the resource-

based view of the ﬁrm, the ﬁrm will focus on its

Part I 91.3

1620 Part I Home, Ofﬁce, and Enterprise Automation

unique, value-adding activities. Successful automation

will enhance these activities and provide (or enhance)

a competitive advantage. Firms can use IT to enhance

ﬁrm value either by lowering costs or leveraging other

ﬁrm-speciﬁc resources that differentiate the ﬁrm from

its competition. Automation is successful here if it low-

ers the bank or credit union’s cost structure or enables

the institution to differentiate its products or services.

Traditional ﬁnance theory deﬁnes success (for an

investment, such as an automation project) as having

a positivenet presentvalue(NPV). Under this approach,

any automation project where (in present value terms)

the expected beneﬁts outweigh the expected cost is suc-

cessful – even if it does not support the bank or credit

union’s strategy.

There are other reasons for automation projects,

including competitive necessity and regulatory compli-

ance. Competitive necessity projects are done to catch

up with the competition. The resulting automation will

not result in a competitive advantage and may not de-

liver a positive NPV (unless the correct baseline is

chosen) but must be done to remain in business. For

these projects, the bank or credit union may go out of

business if the project is not implemented. This should

be the baseline for evaluating the project, not the ﬁrm’s

current environment. Likewise, regulatory compliance

projects confer little beneﬁt to the organization. For

these projects, banksand credit unionsneed to minimize

the cost of the project. The focus of this chapter is on

automation projects that support bank or credit union

strategy.

From manual From packaged

software

From ASP

(outsourcing)

From custom

software

To custom software

To ASP (outsourcing)

To packaged software

To manual

Number of credit unions switching

Original sourcing mode

250

200

150

100

Fig. 91.2 Number of credit unions that switched how they sourced

their core systems

As an example, consider SmallTown Credit Union

(they asked that their real name not be used). Small-

Town Credit Union (STCU) is a community credit

union serving its home county and the surrounding

three counties. It is located in rural, western North

Carolina and historically has served blue-collar work-

ers. It has approximately 15000 members and assets

of US$85million. STCU’s members are neither ﬁnan-

cially nor technically sophisticated. They want good,

personal service, high rates on deposits, and low rates

on loans. To serve these members, STCU invests min-

imally in automation. It uses packaged software for its

core processing and was not an early adopter of Inter-

net banking. It limits the services it offers electronically

and adds new ones only when it is certain of demand by

its current members. STCU’s goals from technology are

primarily to keep operating costs low and only second

to offer new services for their members.

91.3.2 Core Processing

and the Role of Sourcing

The core processing systems at banks and credit unions

support their role as depository institutions and their

role in the payment process. The automation includes

demand deposit accounting, savings accounting, certiﬁ-

cate of deposit and loan accounting systems, as well as

the bank’s general ledger system (every deposit or with-

drawal affects the bank’s balance sheet). These systems

are basic transaction processing systems that track, and

account for,each deposit,withdrawal,and loan payment

made by customers. In addition, they calculate interest

and fees. The demand deposit accounting systems are

coupled with item processing systems that sort checks

and deposits and calculate the totals by account. How

these systems are sourced is a critical driver in their

success.

In 2005, 32 new credit unions and 170 new

(de novo) banks were started. Of the more than 8500

credit unions, almost 500 switched their processing

mode (how their core systems are provided, Fig.91.2).

So, how small banks and credit unions automate their

core processing continues to be of interest and is criti-

cal to their survival. Their ability to meet their strategic

goals depends, to a large degree, on how they automate

their core processing. Although the core systems sup-

port back-of-the-house transactions, they underlie the

bank or credit union’s ability to implement other sys-

tems that enable differentiation.

For banks and credit unions, automation of their

core processing must be linked with their strategic

Part I 91.3

Automation in Financial Services 91.3 Role of Automation in Community Banks and Credit Unions 1621

goals. This automationwill notnecessarily reduce costs.

In the 1960s and 1970s when banks started automating

their core systems, processing capacity increased and

proﬁts for the banking industry as a whole fell [91.6].

For credit unions, based on data ﬁled in their annual

call reports, only 38% of the credit unions that auto-

mated their core processing between 1994 and 2002

loweredtheir operatingcosts ona cost per share account

basis.

Banks and credit unions have three primary sources

for automating their core systems:

1. Develop it and operate it themselves

2. Purchase packaged software and run it themselves

3. Outsource both the software acquisition and oper-

ation to a third party.

The sourcing choice is linked to the alignment between

aﬁrm’sIT strategy and its business strategy [91.17].

Research on outsourcing has addressed the factors that

result in successful outsourcing as well aswhat and why

companies outsource [91.18–21]. The practices that de-

termine management’s perceptionof success aresubject

to debate, and different stakeholders have different per-

ceptions of success. The differences are likely tied to

different views of the ﬁrm’s strategy and the impact of

IT on that strategy.

Production economics [91.1] advocates outsourcing

if the vendor has a production cost advantage. Trans-

action cost economics [91.1, 22] incorporates the cost

of transacting business, including hold-up costs, and

advises against outsourcing activities that are unique

or require asset-speciﬁc resources. The resource-based

view of the ﬁrm [91.23] argues that ﬁrms should in-

vest in their core competencies and outsource the rest.

For small banks and credit unions with limited IT ca-

pabilities, these theories argue for outsourcing core

systems.

For small banks and credit unions much of the ad-

vice related to outsourcing is wrong. For those that

choose to automate their core processing, how they

choose to source their IT function affects their cost

structure. It is related both to the number of products

they offer and the number of channels through they are

available (ability to differentiate). Finally it is tied to

what they pay deposits (value to customers and mem-

bers). Thus, how small banks and credit unions source

their core systems is linked with key strategic decisions,

such as their product and service offerings, their distri-

bution channels, and their pricing.

Outsourcing core systems at small banks and credit

unions is associated with higher operating costs, not

lower operating costs. This contradicts much of the jus-

tiﬁcations given for outsourcing inthe trade press and in

prior research (e.g.,[91.1]) –to lower costs. Thespeciﬁc

vendor chosen also matters, since some software ven-

dors are associated with signiﬁcantly lower costs and

others with signiﬁcantly higher costs.

Instead of cost savings, outsourcing is associated

with more services. Speciﬁcally, ITsourcing affects: the

number of channels (both per member and total), the

number of deposit products offered (total), the number

of loan products offered (per member), and the price of

deposits (both per member and total) [91.24].

This is not surprising. The application service

providers (ASPs) have an incentive to develop new

products (electronic channels) and to promote them to

their clients. They are also able to develop the skills and

processes to help their clients more quickly adopt new

technologies, products, and services. Why credit unions

that use packaged software lag those that use custom-

developed systems is unclear. It could be due to skill

development or to the importance that the credit union’s

executives place on using technology.

The higher costs associated with using an ASP (out-

sourcing) appear to be transferred, at least in part, to

the customer through lower interest rates paid on de-

posits. This is understandable since members use more

deposit-related services (and these services more fre-

quently) than loan-product-related services. They are

willing to pay for the additional channels to support

their deposit products, but unwilling to pay higher loan

rates for a product (loan) that they decide on once and

then require minimal ongoing service.

A number of small ﬁnancial institutions, especially

small credit unions, do their core processing manually.

While manual processing may be considered anachro-

nistic, there are many instances, especially for smaller

organizations, where automation does not make sense

and manual processes are a valid approach. For small

institutions, using manual processes is not more expen-

sive, either in total or on a per-member basis, than using

a computerized system. Identifying the optimal break

point for when to switch from manual to automated

processing is an interesting topic for future research,

but appears to occur by the time an institution acquires

500 members/customers.

Approach

When determining how to successfully automate their

core processing, small banks and credit unions must

ﬁrst determine the basis on which they want to com-

pete. Those that decide to compete on cost need to

Part I 91.3

1622 Part I Home, Ofﬁce, and Enterprise Automation

<1000 1000 to

5000

5000 to

10000

10000 to

25000

25000 to

50000

>50000

1994-CUs with membership in range

<1000 1000 to

5000

5000 to

10000

10000 to

25000

25000 to

50000

>50000

2005-CUs with membership in range

Custom software

ASP (outsourcing)

Package software

Manual processing

Number of credit unions

Credit union size groupings for 1994 and 2005

5000

4500

4000

3500

3000

2500

2000

1500

1000

500

Fig. 91.3 Sourcing of core information systems by credit union size

minimize their overall cost structure (not just their IT

costs). Given this choice, they need to assess whether

to use manual processes or to automate. If they decide

to automate, they need to decide whether to use pack-

aged software or to custom-develop their own software.

The decision to automate or use manual processing

is tied to institution size. Only small credit unions

with fewer than 1200 member use manual process-

ing, and typically they automate well before reaching

this size (Fig.91.3). When automating, the trend is to-

wards using packaged systems. Less than 1% of the

credit unions use custom systems, and only 29 have

switched to custom systems between 2003 and 2005.

Once packaged software is chosen, selecting a vendor

is critical. Again, there is a signiﬁcant difference in op-

erating cost depending on the software package/vendor

chosen.

If the bank or credit union chooses to compete on

service by offering more products and more channels

to deliver these services, then using an ASP is the best

approach. At this point, success becomes a vendor man-

agement issue rather than a technical issue. Choosing

the right vendor and deﬁning the contract are essential.

DifferentASP vendors areassociated with different cost

structures and number of channels. The same vendor

will charge signiﬁcantly different prices based on how

well the bank or credit union negotiates the contract.

Getting a good deal when the initial contract is signed

is critical, since switching costs are high and the vendor

will take advantage of this when it is time to renew the

contract.

91.3.3 Internet Banking

and the Role of Adoption

Bank and credit union’s core systems automate the

back-of-the-house. Banks and credit unions also auto-

mate front-of-the-house processes. These systems (such

as AT Ms and Internet banking) can possibly lower cost

by replacing teller-based transactions. They also pro-

vide new channels for customers to access bank and

credit union services, and they provide additional con-

venience for customers and members. Customer-facing

systems automation can support either a cost-based or

differentiation strategy.

For customer-facing systems, the key to success is

user adoption. ATMs and Internet banking are classic

examples. ATM s were implemented to enable cus-

tomers to conduct ﬁnancial transactions without using

the tellers. The bank and credit unions’ goal was to

reduce costs by eliminating the need for tellers. For

this to happen, customers had to be willing to use the

ATMs. The more customers use these channels and the

more transactions they used them for, the greater the

(expected) beneﬁts for the banks.

Internet banking has also been heralded as provid-

ing a competitive advantage and as being a competitive

necessity [91.15,25,26]. Accordingly, banks spent 15%

of their IT dollars on Internet banking in 2002 [91.27].

Yet, many banks question its proﬁtability [91.28]. In

fact, half of all US banks have no plans to offer Internet

banking, with the most common reason given being an

unclear return on investment [91.14].

Part I 91.3

Automation in Financial Services 91.3 Role of Automation in Community Banks and Credit Unions 1623

Number of adopters

Revenues

Costs

Profits

Fig. 91.4 Impact of Internet banking on costs, revenues,

and proﬁts

Internet banking is deﬁned as a transaction-oriented

system that enables a bank’s customers to engage in

online banking activities. Internet banking enhances

a bank’s offerings by providing its customers with 24/7

access to many of its services. The services available

through Internet banking vary, but typically include

informational account access (view balances and past

transactions), funds transfers among accounts, bill pay-

ment, bill presentment, and some loan application and

approval.

A Booze et al. study [91.26] claims that Internet

banking has an expense rate that is one-third that of tra-

ditional banking with a per-transaction cost of US$0.01

versus US$0.27 for ATM transactions versus US$1.07

for teller-based transactions. For small banks and credit

unions, this cost advantage is unlikely to occur. In-

stead, there is a stylized piecewise-linear relationship

for costs, revenues, and proﬁts driven by the number

of Internet banking users (Fig.91.4). An increase in the

level of consumer adoption increases revenues, costs,

and proﬁts until a saturation level is achieved. Addi-

tional adopters in excess of the saturation level do not

add further to proﬁt, revenue or cost until a second

breakpoint is reached. The cost and proﬁt models expe-

rience a change at this second breakpoint, but revenue

does not. In the cost model, after enough customers

adopt Internet banking, a bank’s costs begin to de-

crease. Correspondingly, proﬁtstrend upwards after this

point.

The initial cost increase comes form a combina-

tion of more transactions, more accounts, and higher

deposits (and therefore higher interest expense). In

particular, banks with Internet banking have a higher

growth rate in the number of accounts (14% higher for

demand deposit accounts (DDAs) and 32.8% higher for

savings accounts). In addition, Internet banking is as-

sociated with a larger average deposit size per account

of about US$800 at a median bank. This indicates that,

while new customers are drawn to the bank, existing

customers also are doing more business with their bank.

This increase in bank activities necessarily increases

costs. Despite this increase in transactions, the actual

volume of transactions performed using Internet bank-

ing relative to more traditional transactions is typically

quite small. Correspondingly, reductions are unlikely in

the ﬁxed infrastructure costs related to traditional trans-

actions (e.g., the number of tellers employed or AT Ms

operated) unless a large number of customers adopt In-

ternet banking.

Improved proﬁts initially result from increased rev-

enues despite increased costs. These improved revenues

are the result of an increase inthe bank’s interest income

through soliciting new customersand gettingmore busi-

ness from previous customers as well as more fee-based

bank services being sold. Previous research ﬁnds that

consumers who use internet banking (IB) are more

lucrative. They generate above-average revenues and

are more proﬁtable [91.29]. In particular, IB customers

have signiﬁcantly higher mortgage and loan balances,

and they use twice the number of other ﬁnancial ser-

vices products [91.14]. The impact of Internet banking

usage on revenue and proﬁt is relatively small and typi-

cally results in an increase of less than 5%.

For the majority of small banks and credit unions

reaching the Internet banking user adoption level break-

points constitutes a major undertaking. For 25% of

the small banks, reaching the breakpoint number of

<1000

members

2772

1000 to

3000

2673

3000 to

6000

1498

6000 to

12000

1072

Exhaust

revenue

benefits

Start

archieving

cost benefits

12000 to

25000

734

25000 to

50000

439

>50000

300

% with

% without

Credit union size category and number of credit unions in the category

100

Fig. 91.5 Percentage of credit unions with internet banking by size

Part I 91.3

1624 Part I Home, Ofﬁce, and Enterprise Automation

adopters requires over half of their customers to adopt

Internet banking. For credit unions, over half do not

have enough members to reach the ﬁrst breakpoint.

For another 25% of small banks, this requires a cus-

tomer adoption level exceeding 25% of its customer

base (Fig.91.5). Doing so also takes time – on aver-

age over 1.5years for those that can and do reach it.

For a larger number of small banks and credit unions

the second breakpoint exceeds their customer base.This

implies that small banks and credit unions have to be

very careful when implementing new customer-facing

technology. To be successful requires getting a signiﬁ-

cant portion of their customer base to adopt.

Approach

When implementing customer-facing automation pro-

jects, banks and credit unions need to take a disciplined

approach. First, they need to determine whether the

project makes sense. This entails assessing the ﬁt with

their strategy. Customer-facing systems will not lower

cost, so the institution must want to differentiate its ser-

vices based on this project. Assuming the project ﬁts

with the strategy, it may still not be feasible if the in-

stitution does not have enough customers or enough

customers who will be interested in using the new ser-

vice. Finally, since the increased revenues come, in part,

from existing customers deepening their relationship

with the bank or credit union, it must offer enough ad-

ditional products and services to take advantage of this

additional business.

If the projectﬁts withthe organization’s strategyand

they have a large enough customer base, they will need

to market the new services effectively. This requires

a different strategy for different target markets. If the

newautomation requires a signiﬁcant change in how the

customer uses the bank or credit union services (as was

the case with AT Ms and Internet banking), they should

consider paying the customer to use the service. If cus-

tomers will be paid to use the service, it must be done

carefully. Banks and credit unions that paid customers

to use AT Ms saw increased adoption. Some banks and

credit unions that paid customers to create Internet bank

accounts saw people createaccounts, but not necessarily

use them.

If the automation project is innovative, the cus-

tomers will need to be trained to use the new system.

Younger customers may be able to learn online, but

older customers and those whose primary channel is the

branch will need more personal training, probablyat the

branch. Finally, the banks and credit unions need to lis-

ten to their customers who are not using the new service

or channel to determine why, and then to either mod-

ify the service/channel or update their marketing and

training materials.

Even with an effective marketing program and

a good training process, thebanks and creditunions will

face a customer adoption process. Customer adoption

will follow an S-curve. Effective marketing and train-

ing may make the curve steeper, but identifying and

supporting early adopters is critical. Early adopters of

an innovation typically have the most interest and will-

ingness to pay for the innovation. Later adopters are

typically less inclined toward the innovation. Relative

to new services, this implies that customers who adopt

them early on are likely to be the most proﬁtable, and as

more and more people adopt, the incremental revenue

per customer will diminish.

Sourcing Impact

As with core systems, Internet banking, and future

customer-facing automation projects, how the automa-

tion project is sourced impacts success. The sourcing

choice changes the cost structure for the service. It

affects the implementation timing and potentially sup-

ports the customer adoption process.

When the banks and credit unions use an Internet

banking service provider, their variable transaction cost

structure is different from that for banks that develop

their own Internet banking system. To implement Inter-

net banking, there is still a ﬁxed, but small, one-time

cost for the system in the US$20000–40000 range.

Once the system is installed, there is a nominal ﬁxed

monthly fee (less than US$1000) plus a fee for each

Internet banking customer as well as fees for various

transactions. Thetotal fee for each customer depends on

the services the customer is signed up for (e.g., bill-pay)

and how much historical information the bank retains

(i.e., itcosts the bank more ifthe bank lets itscustomers

view deposits and withdrawals from 2years ago than

if the maximum a customer can go back is 6months).

This fee can range from US$2 to US$ 15 per customer

per month. Additional fees also are charged for most

transactions, including signing up a new Internet bank-

ing customer and for a customer making a payment.

Enquiries are free. This cost structure makes most of

the bank’s costs variable. This lowers the risk of imple-

menting Internet banking, but also makes lowering the

bank’s overall operating costs more difﬁcult. The op-

erational risks typically associated with innovation are

mitigated by the use of a vendor, but banks that use this

service are unable to physically distinguish themselves

from other banks using the same service.

Part I 91.3