I



OPERATIONS IN FINANCIAL SERVICES

Prof. M. Pinedo

THE REASONS BEHIND THE UNITED STATES RELIANCE IN CHECKS: A PROPOSAL TOWARDS COST EFFICIENCY

Gorka Briones

Cristina San Jose

May 2005

THIS PAPER HAS BEEN WRITTEN WITH PUBLIC INFORMATION

Check processing in the 1960s

Nodaway Valley Bank, Northwestern Missouri

SUMMARY

In the United States, retail noncash payments are usually made using checks, credit cards, debit cards, and an electronic payment system called Automated Clearing House (ACH). Consumers, businesses and government entities made 70+ billion retail noncash payments in 2000. The total value of these payments in 2000 was $45+ trillion, more than four times the US GDP for that year. The Federal Reserve processes about 40% of the US checks while the rest are processed by big banks and private clearinghouses.

This paper analyzes the reasons behind the low penetration of electronic payment instruments in the US, and the potential benefits and likelihood of a change in the payment instruments mix. We have focused our analysis in Checks and ACH payments, as suggested by comparing the US payment mix with the one in place in other developed economies. Our conclusion is that compatibility issues between the Check and the ACH systems, as well as a lack of incentives for retail customers and banks to switch to ACH payments are in the roots of the problem.

To analyze the potential benefits of an increase in the penetration of ACH transactions, we assess the cost of each network. Our analysis shows that because of its current mix of payment instruments, the US system might be incurring in much higher costs that its European counterparts, as an ACH payment is more than 50% cheaper than paper based instruments.

To analyze the likelihood of a natural increase in the demand of ACH transactions we studied the various incentives for stakeholders in the network (retail customers, business customers and banks) to switch to ACH payments. On the demand side, we found that there are no strong incentives for retail customers to switch to ACH, while both cost and revenue incentives are pushing business customers towards ACH. On the supply side, the importance of checking related revenues, the existence of network externalities and the fragmentation of the industry are preventing non ACH enabled banks to offer the service. Then, we analyze and propose several incentives to boost ACH as a more cost efficient payment system.

Additionally, we outline the DEA analysis for Check processing in a fully compatible payment system, as the asymmetric treatment of float and non sufficient fund fees create different performance indicators for the different instruments. The trade off between productivity and quality is highlighted in the analysis.

INDEX

I. TRENDS IN THE USE OF CHECKS IN THE US

Summary 8

a) Decline in the number of checks 8

b) Downsizing of check processing facilities 9

c) Check 21 Act 10

II. INTERNATIONAL COMPARISON

Summary 11

a) US dependence on checks 12

b) Situation of credit/debit cards 13

c) ACH, Credit Transfers and Direct Debits 15

III. THE CHECK 21 ENABLED NETWORK VS. THE ACH NETWORK

Summary 17

a) Check 21-enabled checking network 18

b) ACH network 19

c) Compatibility of Paper Checks’ and Image Checks’ Network 20

d) Compatibility of Paper/Image Checks’ Network and ACH 20

e) Detailed Compatibility Overview 21

IV. COST IMPLICATIONS OF THE CURRENT US PAYMENT MIX

Summary 22

a) Processing costs 22

b) Cost comparison across networks 24

V. CURRENT INCENTIVES FOR RETAIL AND BUSINESS CUSTOMERS FOR CHOOSING EACH NETWORK

Summary 25

a) Check float vs. the predictability of ACH settlement 25

b) The perceived price of check transactions 27

c) Convenience 28

VI. CURRENT INCENTIVES FOR BANKS FOR CHOOSING EACH NETWORK

Summary 30

a) The importance of checking related revenues 30

b) The existence of Network Externalities 31

c) The fragmentation of the US banking industry 33 33

VII. SUMMARY OF CURRENT INCENTIVES FOR CHOOSING EACH NETWORK 35

VIII. POTENTIAL POLICIES TO INCREASE DEMAND FOR ACH TRANSACTIONS

Summary 36

a) The need for additional incentives 36

b) What are the potential incentives? 39

c) Incentives proposal 40

IX. DEA ANALYSIS OF CHECK PROCESSING IN FULLY COMPATIBLE PAYMENT NETWORKS

Summary 42

a) Check processing flow in fully compatible networks 43

b) DEA Analysis of Step 2: Bank’s Central Processing Center (Collection from Branches/Sorting/Credit Customer’s Account/Delivery to the Clearing House) 44

c) DEA Analysis of Step 3: Clearing House (Collection/Sorting/Delivery to the Paying Bank) 46

d) DEA Analysis of Step 4: Paying Bank (Collection/Sorting/Check Availability of Funds/Debit Customer’s Account or Return Check to Collecting Bank) 48

BIBLIOGRAPHY 53

I. TRENDS IN THE USE OF CHECKS IN THE US

Recent years have seen dramatic increases in the use of electronic modes of payment in the US (a)

As check volumes decline, unitary costs go up, due to the fixed costs represented by processing centers. Both banks and the Federal Reserve are attempting to reduce these fixed costs by using part-time staff and consolidating the processing centers (b) However, given that these centers are built to process large volumes, any consolidation will necessarily remove processing capacity in large chunks, rather than matching the slow but steady decline in volumes.

With banks reporting checks declining from 3-8% a year, the need to move to electronic check processing has become acute. The October 28, 2004 effective date of the Check Truncation Act (Check 21 Act) has created a deadline on which banks and industry associations are focused (c).

a) Decline in the number of checks

The number of noncash payments in the United States has grown since 2000 at a 3.8% annual rate. Checks are the only payment instrument being used less frequently now than three years ago.

[pic]

Electronic payments in 2003, including those made by credit and debit cards, exceeded check payments for the first time in the United States, according to the Federal Reserve. The Fed also estimates that credit cards and debit cards will both surpass checks in terms of total annual transactions by 2007. The number of electronic payment transactions totaled 44.5 billion in 2003, while the number of checks paid totaled 36.7 billion, down from 50 billion in 1995.

Check payments have decreased at an annual average rate of 4% from 2000 to 2003. As for electronic payment methods, they have increased at an average annual rate of 13% for the same period.

According to a survey by the American Bankers Association, consumers are increasingly shifting to electronic methods to pay regular bills. In 2003, 40% of bills were covered by either automatic deductions from bank accounts or through online bill payment. Electronic payment accounted only for 15% of the bills in 1979.

“People is becoming increasingly comfortable doing things online”, says a spokesperson from Bank of America, “Customers might start out looking at their monthly statement, then they might transfer money from savings to checking and then they might start paying a few bills. And once they try it, they see how easy it is and they love the convenience”

b) Downsizing of check processing facilities

As checks are phased-out, their earlier processing and sorting scale economies work in reverse and raise unit costs as volumes fall. In February 2003, even the Fed reported that “mainly because of declining check volumes”, its own financial performance had suffered over the past year. As a result of its dismal performance, the Fed has announced the downsizing of its check processing operations.

“In August 2004, the Fed launched a second round of consolidation that will reduce the number of processing centers to 23 by 2006, from 32 at the end of 2004 and 45 in 2003, when it began restructuring its check operations. This latest round will cut a net of 270 jobs, on top of the 400 jobs cut in the first round [1]”

c) Check 21 Act

The Check 21 Act took effect in October 2004, and it is designed to be a flexible legislation, intended to speed the move toward electronic exchange away from physical transport of paper checks, facilitating check truncation. Interbank check truncation was pioneered in Germany, with the UK, France, Hong Kong and Singapore following close.

Check 21 gives banks the legal framework to accept digitized images of checks instead of hard-copy paper checks. The legislation was enacted for two reasons: as a post-Sept. 11 safety net for lost checks through business interruption and to cut costs for banks physically transporting paper checks.

Check 21 does not mandate that banks present checks electronically, but creates greater flexibility, which allows banks to choose between paper, image or a combination of both. Some banks feel, however, that Check 21 diverts attention away from other electronic payment initiatives, and does not deter customers from writing costly checks.

It is expected that Check 21 will reduce costs and provide increased flexibility to banks and consumers. Additionally, Check 21 will reduce overall risk, but it will also increase the vulnerability in certain areas, such as networks and IT systems, at least in the short run. As checks will clear sooner, the risk that a check will bounce back if funds are not in the account when the check is first written will also increased.

II. INTERNATIONAL COMPARISON

In spite of the steep decline in paper-based payment instruments in the US, the country is still heavily dependant on checks when compared to other developed economies like Japan or the European Union. The US is the country with the lowest penetration of electronic forms of payment relative to the volume of cashless transactions. Electronic forms of payment include debit and credit cards as well as ACH, credit transfers and direct debits. Additionally, the US is the country that writes more checks per capita (a)

If we focus in electronic payment methods only, the US is well ahead the European average in credit/debit card penetration relative to the total volume of cashless transactions. Only Denmark, Greece, Luxembourg and Portugal have higher penetration rates for these instruments. (b)

Therefore, the differences in check usage between Europe and the US must be mainly caused by electronic payment instruments other than credit/debit cards: Credit Transfers and Direct Debits. In fact, the rate of penetration of those instruments in Europe is 5 times bigger than ACH penetration in the US. (c)

a) US dependence on checks

In spite of the recent decline, the extent to which electronic forms of payment have substituted for checks in the US is less than what is often supposed. Among developed countries[2], the US remains the most dependent on the use of checks.

The US wrote an average of 125 checks per capita in 2003, which is more than 5 times the amount of checks written by the European Union average in 2001 (22 checks per capita). If we take into account the fact that this figure has been decreasing at an annual rate of 5% in the last five years (the average European wrote 27 checks in 1997 vs. 22 in 2001) the gap with the US would increase.

Among the EU members, France, UK, Portugal and Ireland are the heaviest check users, but still well below the US average. French write 6 checks a month, Britons write 4 and Portuguese and Irish write 2, while Americans write 10 checks a month.

Belgians, Danish, Germans, Spaniards, Italians and Austrians write less than 1 check a month, and Greeks, Luxemburguese, Dutch, Finesse, Swede and Japanese write no checks at all.

According to the share of checks in noncash transactions volume, the US appears as the most check dependant country, whit a penetration of checks (45% of total transactions) 3 times the European average (15%). Again, if we had available 2003 data for Europe the gap with the US may be wider, as check penetration has decreased 8 p.p in the last five years (check penetration in the EU was 23% in 1997 vs. 25% in 2001)

Other countries with a high penetration of checks are France (35%), UK (24%), Portugal (27%), Ireland (30%) and Italy (20%), all below the US average. These countries have been experiencing a steep decline in the usage of checks, with huge penetration loses between 1997 and 2001. In this period, Ireland lost 31 p.p., Portugal lost 15 p.p., France and the UK lost 12 p.p and Italy lost 10 p.p.

b) Situation of credit/debit cards

The US reliance on checks is not the result of a lower penetration of credit/debit cards as payment instruments. In fact, cards penetration in the US volume of transactions (42% in 2003[3]) is much higher than the European average (no data for Japan are available)

The European average was 29% in 2001[4]. In the US, debit card transactions alone account for over half of all growth in electronic payments.

Especially interesting are the cases of Germany, Spain and Austria, were credit/debit card transactions account only for 11%, 26% and 15% respectively of all transactions (vs. 42% in the US) and still have very low rates of check penetration (2%, 7% and 1% vs 45% in the US)

c) ACH, Credit Transfers and Direct Debits

From the data above we conclude that the differences in check usage between Europe and the US are mainly caused by electronic payment instruments other than credit/debit cards.There are three other instruments that we should consider: ACH in the US and Credit Transfers and Direct Debits in Europe.

ACH[5]:

The ACH is an automated system for exchanging electronic funds drawn against one another. The ACH system supports both debit (e.g., automatic payment of mortgages) and credit (e.g., direct deposit of payroll) transactions. Unlike the wire transfer, it is usually used to process higher volumes or small-dollar payments for settlement issues within 1 to 2 business days.

Developed in the 1970s, the ACH system was a banking industry attempt to replace the paper check. A collection of 32 regional electronic inter-bank networks is used to process the transactions electronically with a guaranteed one-day bank collection float.

Credit Transfer and Direct Debit[6]:

A Credit Transfer is a payment order or possibly a sequence of payment orders

made for the purpose of placing funds at the disposal of the beneficiary, like the deposit of an employee’s payroll. Both the payment instructions and the funds described therein move from the bank of the payer/originator to the bank of the beneficiary, possibly via several other banks as intermediaries or more than one credit transfer system.

A Direct Debit is a pre-authorized debit on the payer’s account by the payee, typically used to pay periodical bills such as utilities, telephone or rent.

We are going to compare the penetration of Credit Transfer and Direct Debit in Europe with the penetration of ACH in the US as its closest substitute.

Germany, Spain and Austria, the countries with the lower penetration on both checks and credit/debit cards have the highest penetration of Credit Transfers/Direct debits, with 86%, 66% and 84% in 2001, compared with a penetration of 11% in the US in 2003. The European average is also well above the US, with a penetration of 56% in 2001. Actually, the average usage of Credit Transfer/Direct Debit has decreased slightly in Europe, falling from 63% in 1997 to 56% in 2001 as a result of the increase in credit/debit cards usage.[7]

III. THE CHECK 21 ENABLED NETWORK VS. THE ACH NETWORK

In this Chapter, we will address the compatibility among the Check and the ACH Network. As it is not mandatory for banks to accept ACH transactions, compatibility issues arise. Additionally, a recent piece of legislation (Check 21 Act) has added an extra layer of complexity, as the Check 21 forces banks to accept truncated checks from the Clearinghouses. Check 21 does not mandate that banks present checks electronically, but creates greater flexibility, allowing banks to choose between paper, image or a combination of both. A detailed overview of the different components of the Check 21-enabled Network and the ACH Network can be found in (a) and (b)

To analyze compatibility, first we analyze the compatibility between paper and image checks inside the Check 21-enabled Network in (c). Then, we analyze compatibility between the Check and ACH Networks, as for paper initiated ACH payments both banks need to work with ACH systems (d)

As a summary, we provide a detailed overview of the different compatibility issues among the various networks in (e)

a) Check 21-enabled checking network

[pic]

b) ACH network

c) Compatibility of Paper Checks’ and Image Checks’ Network

The Check 21 Act legally enforces compatibility between paper checks and image checks, by forcing the Paying Bank to accept substitute checks. Under the Check 21 Act, the Federal Reserve may send to the Paying Bank either an image or a printed substitute check. Banks can choose whether or not to accept and create images, but they must accept substitute checks.

Clearing Houses are in charge of melding both networks into one. This way of enforcing compatibility will result in two outcomes. First of all, customers will have to get used to not having their cancelled checks back. Secondly, every bank will have to invest in Check 21 Substitute Check Reading technologies to be able to process the files sent to them by the Federal Reserve. Otherwise, they will have to deal with them manually. The investment in Check 21 Image Creating technologies remains optional, so that every bank will be able to keep on sending paper checks.

d) Compatibility of Paper/Image Checks’ Network and ACH

An ACH transaction can be initiated either by a paper check or by an electronic payment order. When initiated by paper check at the Point of Sale/Lockbox, the check is converted into an electronic file and either given back to the Paying Party (Receiver) or destroyed. When initiated electronically, the Paying Party authorizes the Collecting Party to debit his account (online payment of utility bills, telephone). In both cases the Paying Party will receive an ACH report of the transaction.

The Federal Reserve is able to clear ACH transactions, but the acceptance/issuance of ACH files is not mandatory for banks. Both networks remain incompatible, as the Federal Reserve can not collapse both networks the way it does with paper and image checks.

Therefore pure electronic payments only happen when every party has voluntarily agreed to use them: the paying customer (who initiates the transaction or allows it) the collecting bank and the receiving bank (who both work with ACH systems). For paper initiated electronic payments both banks need to work with ACH systems.

e) Detailed Compatibility Overview

IV. COST IMPLICATIONS OF THE CURRENT US PAYMENT MIX.

In this Chapter we analyze the cost implications of the current payment situation in the US. Due to its current mix of payment instruments, the US system might be incurring in much higher costs that its European counterparts, as an ACH payment is more than 50% cheaper than a paper based payment. In Europe,“econometric analysis suggests that the shift to electronic payments over 1987-1999 may have reduced the cost of making a payment by 45%”[8](a)

In a cost comparison across networks, ACH transactions are the cheapest for the overall system, while paper and image transaction have higher processing costs (b)

a) Processing costs

As a general rule of thumb, an electronic payment costs only from one-third to one-half as much as a paper-based payment:

[pic]

Ranging from 7 cents/transaction in Spain 2001 to 18 cents/transaction as reported by the Federal Reserve in 2000, ACH/Credit Transfer/Direct Debit is by far the cheapest payment instrument other than cash for both banks and retailers. Because of its current mix of payment instruments, the US system might be incurring in much higher costs that its European counterparts.

According to David Humphrey of the Federal Reserve Bank of Philadelphia: “For 12 European countries, the share of electronic payments in total non-cash transactions rose from 43% in 1987 to 79% in 1999 to 85% in 2001.Econometric analysis suggests that the shift to electronic payments over 1987-1999 may have reduced the cost of making a payment by 45%. This shift, along with the large expansion of ATMs in place of expensive branch offices, is the main reason why the ratio of bank operating cost to total asset value in Europe has fallen by 24% over the same period. A few European countries have already largely shifted to electronic payments and are realizing the cost savings involved. Others, along with the U.S., are at an earlier stage.”

b) Cost comparison across networks

Currently, a Collecting Bank bears the bulk of the costs associated with clearing the check. In other words, a Paying Bank is under no legal obligation to share a Collecting Bank's costs of collecting checks drawn by the Paying Bank's depositors. Increased penetration of the ACH Network would generate positive savings to the US payment system.

V. CURRENT INCENTIVES FOR RETAIL AND BUSINESS CUSTOMERS FOR CHOOSING EACH NETWORK

There is a variety of reasons to explain the prevalence of checks vs. ACH payments in the US. Here we are going to analyze this issue from both the perspective of customers and banks.

From the customer’s point of view, the possibility of playing the check float vs. the predictability of ACH payments (a) as well as the perceived price of check transactions (b) partially explain their preference for checks. Additionally, checks are perceived as convenient by the US customer. On the one hand, the typical US customer likes to keep checks as the record of his transactions. Yet on the other hand, more than 15% of the US population is “unbanked”, relying on Wells Fargo checks as their only payment instrument available (c)

The Check 21 Act will diminish the relevance of some of these features, thus eliminating some of the incentives to use checks (d)

a) Check float vs. the predictability of ACH payment

The first reason to explain the US reliance on checks is the check float, which is the interest earned by a check writer between the time a check is received as payment and the time the payment is settled. This time lag allows people to “play the float”, giving themselves short-term, interest-free loans There are several key features of the U.S. check payment system that contribute to the creation of float.

First of all, the US check payment system is highly decentralized, with more than 10,000 banks, savings institutions and credit unions. Therefore the probability that a check is being written between customers of the same institution (“on-us” checks) is small, only 24% in 2004[9]. To settle on-us checks, the bank would just debit the payor’s account and credit the payee’s. However, to settle not on-us checks the bank needs to interact with the Federal Reserve System/Private clearinghouse and another bank. This process means that check clearing can often be subject to several days.

This fact, which constitutes an advantage for retail customers (individuals), becomes a disadvantage for business customers. Business customers are usually net receptors of checks (telecom companies, utilities) and therefore they lose the extra interest their customers are making.

Secondly, there is a legal requirement that checks presented for payment be paid at par, or full value. The requirement to pay at par does not vary with the amount of time it takes to present a given check. This feature of the check payment system implicitly defines a zero interest rate for check funds in the process of collection.

In the case of ACH transactions, the timing of settlement is flexible, but it is also precisely controllable by the payor and payee. The predictability of the clearing and settlement process reduces the scope for people to undertake actions in order to manipulate float since such actions would immediately be observable to the other parties involved in the transaction. Hence, float is generally not an issue for these types of transactions.

According to the UCC article 4A-401, for funds transfers (ACH payments) payment date is defined as:

"Payment date of a payment order means the day on which the amount of the order is payable to the beneficiary by the beneficiary's bank.  The payment date may be determined by instruction of the sender but cannot be earlier than the day the order is received by the beneficiary's bank and, unless otherwise determined, is the day the order is received by the beneficiary's bank.”

Additionally, customers’ perception about the benefits from check float may be overstated. Recent reductions in check-processing times have vastly reduced the value of float for the average checkwriter. “The clearing process has been becoming faster thanks to the following initiatives[10]

• The use of magnetic ink on checks for identifying the customer account and paying bank to be debited

• The standardization of the dimensions and thickness of a check

• The requirement that bank endorsement information be placed where it can be easily read (to facilitate return item processing)

• The development of ever faster check reader/sorter machines (made possible by the use of magnetic ink and standard check dimensions)”

b) The perceived price of check transactions

The second reason for checks prevalence in the US is the way its price is perceived among customers. The pricing of payment instruments can strongly affect the usage mix of these instruments. In Norway, paper-based instruments suffered a dramatic decline after the banks agreed to price every transaction according to its unit marginal cost. The share of paper-based transactions fell from 90% in 1987 to 40% in 1996[11].

In the US, “while retailers and businesses usually pay a price per transaction they make or receive through their bank, consumers typically choose to hold a required minimum balance and/or pay fixed monthly account fees rather than pay a fee per transaction”[12]. In the period from 1997 to 2001, the percentage of banks and savings associations offering monthly fee-only checking accounts in the US increased from 29% to 38%.

These requirements and fees reflect the bank’s estimated average price for its checking services taking into account the various customer’s usage profiles. But because of these flat or semiflat fees, customer’s perceive check transactions as having a zero marginal cost.

To reinforce this perception banks have increased only moderately monthly fees and minimum opening balances of checking accounts11. With banks reporting checks declining from 3-8% a year, this price moderation barely reflects the increases in unit cost.

[pic]

For business customers, checks are perceived as more costly than electronic based payment instruments. As an example, the US Government is heavily promoting the electronic payment of Social Security benefits and tax refunds.

c) Convenience

In general, US customers perceive checks as convenient. First of all, checks give consumer a “sense of control”

• Checks allow customers to see the signature of their debtors/creditors

• Checks allow customers like to keep physival checks for audits

• Checks partially relieve the customers’ distress for transactions errors

Secondly, above 15% of the US population remains “unbanked”. These people have to relay on Money Transfers or Wells Fargo Checks as their only payment instruments available.

For business customers, checks are less convenient because of the increased complexity of Working Capital Management.

d) Check 21’s disincentives for retail customers

First of all, the Check 21 Act will reduce the float for customers, as it will take an average of only 1 day for checks to clear. The incentive for float playing will decrease dramatically, as increased probability of overdraft and bounced-check fee charges would dilute the benefits from floating.

Secondly, customers will no longer receive their cancelled checks back. After taking a digital image, the bank will destroy the original check. If necessary, a substitute check may be created. Customers will receive substitute checks or printed non substitute checks, eliminating one of paper checks’ convenient features.

VI. CURRENT INCENTIVES FOR BANKS FOR CHOOSING EACH NETWORK

Also bank’s have many reasons to support checks in spite of its higher processing costs. Checking is an important revenue source for retail banks, with annual collections of more than $4Bn of checking related fees. Specially relevant are the non-sufficient fund related fees, which account for 30 to 65% of the total bank’s consumer checks revenues. Because of its non-recurrent character, customers are quite inelastic to NSF fees, which have been growing at more than 5% annually in the period 1997-2002. ACH payments would reduce fee income for the banks, specially NSF fees as real time verification of funds would be posible for certain transactions. From the banks’ perspective, there is a trade off between the revenue losses and the cost savings of adopting ACH (a)

As with any new technology, there are network externalities in the ACH system. Consumers and banks will be more likely to adopt ACH the more other parties do. Therefore the still very low ACH penetration rate in the US may be discouraging banks from adopting ACH (b)

Also the fragmentation of the US financial system contributes to the slow adoption of ACH payments. This explains the earlier adoption of electronic payment systems in Europe, were the industry is much more concentrated. ACH adoption is higher for large markets and in highly concentrated markets. Banks with few local competitors are more likely to invest in ACH (c)

a) The importance of checking-related revenues

From the perspective of the banks, checking is a valuable revenue source. According to IBM Consulting and Celent, only the total annual fee revenues generated by the consumer check payment value chain are US$6.7B, excluding clearing, mailing and transport services. Banks capture more than 70% of this amount. Banks' consumer check revenues come from

• fees charged to merchants for depositing checks

• lockbox fees charged to billers for collecting payment by checks

• checkbook sales and non-sufficient funds related fees charged to consumers

Non-sufficient fund related (NSF) fees are the largest component of the banks' revenue stream, accounting for 30-65% of banks' consumer check revenues. NSF fees also drive the highest margins, with a 90% gross profit margin.

All the fees associated with special actions, such as stop-payment orders and NSF checks, have risen significantly and by substantially more than inflation over the 1997-2002 period[13]. Because of the non-recurrent character of these fees, customers are quite inelastic to these increases.

[pic]

With some electronic payment instruments such as Direct Debit, banks can check in real time that the accounts are valid and the funds are available, therefore missing the NSF fee. In general, banks can charge customers more for processing checks than for processing ACH payments. Thus, the substitution of checks by ACH payments would imply a considerable reduction in the bank’s top line.

b) The existence of network externalities

The substitution of checks by ACH payments would decrease marginal costs for banks. From the banks’ perspective, there is a trade off between the revenue losses and the cost savings of adopting ACH. The change in profits from adopting the ACH technology is the number of ACH transactions times the difference in margins, minus the fixed cost of adoption.

For JP Morgan Chase & Co., the benefits of ACH payments clearly exceed checks’. Leonard J. Heckwolf, the senior vice president and product head for ACH and retail lockbox operations at J.P. Morgan, the largest ACH originating bank, said that “the long-term cost reductions that come from using the ACH system are far more important to a bank’s bottom line than the potential declines in fees and float”

JP Morgan’s strong endorsement of ACH payments is better understood in the context of networks economics. ACH payments form a network in which consumers and banks exchange payments with other consumers and banks that also accept the technology. Therefore, electronic payments systems are useful only if other entities also use them. ACH forms a network because both the originating bank and the receiving bank or its designated correspondents must accept ACH to complete the transaction.

“As there is a fixed cost for consumers and banks to adopt ACH, then when one consumer or bank adopts ACH, other ACH adopters are better off, since they can exchange payments with one more party. Therefore, consumers and banks will be more likely to adopt ACH the more other parties do. Moreover, because ACH usage is technologically intensive, additional users may improve “user friendliness,” which may lower the costs of usage to other consumers. Both effects lead to an externality in which one user’s increased usage of ACH makes other users better off[14]”

Because of the low adoption of ACH payments in the US (11% of all the cashless transactions in 2003) externalities in the ACH network are still very small. This low penetration rates are discouraging banks from adopting ACH. Large/Medium banks like JP Morgan can afford ACH adoption in spite of the technology’s low penetration, as they can spread the fixed cost of adoption over more customers.

US Bank’s ACH adoption by asset size13

[pic]

The data above show the consistently higher penetration rates of ACH technology in large banks. More updated data are not available, altough the trend is likely to have been sustained: ACH penetration in 1997 was 5% vs 11% in 2003. Most likely, 100% of large banks will be ACH enabled by the end of 2005.

c) The fragmentation of the US banking industry

The high fragmentation of the US financial system may be also contributing to the low pace of ACH adoption. Banks in large markets and banks with few local competitors are more likely to invest in ACH, since they are not waiting for others to adopt ACH before they adopt it themselves. This would be consistent with the higher penetration of ACH payments in Europe, as the European banking industry is much more concentrated. In general, the more concentrated the area, the higher the adoption of ACH.

In the FRBSF Economic Letter Number 2002-27, Gowrisankaran shows the influence of market size and market concentration in bank’s adoption of ACH technology in the US:

[pic]

Gowrisankaran divides the US in metropolitan areas, and estimates the banking concentration using the Herfindahl Index. The more concentrated the market (high Herfindahl Index) the higher the penetration of ACH technology. Additionally, the larger the market, the higher the ACH adoption.

VII. SUMMARY OF CURRENT INCENTIVES FOR CHOSING EACH NETWORK

VIII. POTENTIAL POLICIES TO INCREASE DEMAND FOR ACH TRANSACTIONS

In this Chapter we analyze the need for additional incentives to increase the demand of ACH payment services, in light of the current incentives structure discussed in Chapters VI and VII. The analysis is done from the perspective of both the demand and the supply side (a)

Then, we list all the potential incentives that could be used for retail customers and banks in the Check-only network, with a reference to successful incentives implemented abroad (b). Additionally, we evaluate the feasibility of the various incentives we have found to elaborate a final proposal (c)

a) The need for additional incentives

ACH networks would be less costly for the US payment system. However, the current incentives structure contributes to the low penetration of ACH payments in the US, in both the demand and the supply side.

First, we will analyze the demand side for ACH transactions. For the average Retail Customers, electronic payments have no clear advantages vs. traditional paper checks. For Business Customers, the advantages for ACH payments are both in the revenue (float) and cost side (processing). Because of this asymmetry, recent efforts by Business Customers to migrate Retail Customers to online payments have had a limited success only.

[pic]

Secondly, we will analyze the supply side. We discussed in Chapter VII the lack of incentives for small-sized banks and banks in highly fragmented/smaller markets to adopt ACH technologies, thus keeping their network private. This situation has two effects in the system. First, it prevents their Retail Customers to use ACH even if interested. Second, the overall Payment Network remains incompatible, forcing other players to transact in costlier ways even if ready to transact electronically.

As described in Chapter IV, an increase in the demand of ACH would lead these institutions to adopt compatibility becoming ACH ready, even in the absence of other incentive or binding agreements. For this to happen, the ACH demand should be as large as the demand for check transactions, which today are 4 times the number of ACH transactions. However, due to the lack of incentives for Retail Customers, this large increase in demand is unlikely to happen. Additionally, as smaller commercial banks generally have a higher share of heavy check transactors Retail Customers in their customer base, it is unlikely that they will face customers’ pressure to adopt ACH technologies.

b) What are the potential incentives?

Market incentives are already in place for both Business Customers and Net Collecting Banks, as the second may lose the first if not ACH enabled. However, this is not the case for non ACH-enabled Net Paying Banks: given that Retail Customers have no strong incentives for ACH, these banks can remain Check-only. Then, to boost ACH demand, we would need additional incentives for both Retail Customers and non ACH-enabled Net Paying banks.

Several incentives to boost retail demand for electronic payments have been put in place internationally:

• Canadian banks eliminated the check float for both retail and business customers. “Canadian banks give payees same - day funds availability for deposited checks and charge firms for the check float they create. The central bank also backdates interbank check settlements by one day to effectively eliminate interbank float”[15] (Float incentive)

• In Norway, paper-based instruments suffered a dramatic decline once the banks accorded to price every transaction according to its unit marginal cost. The share of paper-based transactions fell from 90% in 1987 to 40% in 1996[16]. Thus, charging US customers with a Fee per check might reduce the appeal of paper instruments (Fee incentive)

As for banks, we showed in Chapter VII that small-sized banks in fragmented and small markets are the ones with less externality benefits from ACH adoption. An ACH adoption subsidy for these players might be enough to encourage them to become ACH compatible, thus increasing the appeal for compatibility among other check-only players (Subsidy incentive)

c) Incentives proposal

The fragmentation of the US banking sector compared with the Canadian and Norwegian makes it difficult to implement both the Float and the Fee incentives. Additionally, given that there are Check-only banks, introducing the Fee incentive would punish customers that may not have other alternative for banking. Therefore a subsidy for smaller banks would be our final proposal. Additionally, once the Payments Network becomes compatible, the Fee incentive may be reconsidered to further boost ACH demand.

IX. DEA ANALYSIS OF CHECK PROCESSING IN FULLY COMPATIBLE PAYMENT NETWORKS

In this Chapter we will outline the DEA analysis for Check processing in a fully compatible payment system, as the asymmetric treatment of float and non sufficient fund fees create different performance indicators for the different instruments.

First, we draw the process flow diagram corresponding to the compatible system, where paper checks, image checks and ACH transactions could be processed by the same parties (a)

As there is a trade off between productivity and quality, we provide inputs and outputs for both.We analyze the Bank’s Central Processing Center in (b), the Clearing House in (c) and the Paying Bank in (d)

a) Check processing flow in fully compatible networks

In a fully compatible payment network, Paper Checks could be handled in three different ways:

• Truncation: Under the Check 21 Act, paper checks could be converted into image files to be transmitted electronically

• Conversion: Because of compatibility, paper checks could be converted into ACH files to be transmitted electronically

• Traditional: Traditional processing of physical paper checks

To understand the presence of paper checks in the process it is necessary to mention two things. First of all, the truncation of checks (the transformation of paper checks into image checks) is not mandatory. The only mandatory implication of the Check 21 Act for banks is the reception of substitute checks from the Federal Reserve. Secondly, the conversion of paper checks into ACH files requires the agreement of the customers. Finally, not every paper check can be converted into an ACH file (i.e, cashier’s checks can not be converted). Therefore, a certain amount of paper checks will always have to be handled traditionally.

There are 4 major steps in the process, each of them involving several transactions. Some of the transactions in the diagram involve the transfer of funds (blue) and the others involve the handling of either paper/image checks (red) or paper initiated ACH files (red). Our analysis is going to focus in the handling of checks rather than in the transfer of funds. We exclude Step 1 from the analysis because of major redundancies with Step 2.

b) DEA Analysis of Step 2: Bank’s Central Processing Center (Collection from Branches/Sorting/Credit Customer’s Account/Delivery to the Clearing House)

Productivity:

Because of the asymmetry in float treatment for Paper/Truncated checks and ACH files, the bank should consider them differently.

• Paper/Truncated checks: From the Collecting Bank’s perspective, productivity is inversely related to the amount of time it takes them to deliver the checks to the Clearing House. Because of the float, every delay means that the Collecting bank is losing spread. Different “delay tolerances” may exist according to the value of the check.

• ACH Files: ACH is float neutral, and therefore time considerations do not account for ACH files

Therefore, we will have to consider 2 outputs for the DEA analysis:

• “processing days/check of size(i)”, being size(i) different values of paper/truncated checks for which the bank may have different tolerance levels

• “number of ACH files processed”

As for the inputs, productivity will heavily depend on the “# of entry points” the bank’s processing center is collecting checks from (generally the number of bank branches it serves) as well as the “total # of items (checks and ACH files)” it collects. Additionally, we would have to consider the “Mix (%) of paper/image/ACH” and “% of paper/image checks of size(i)” the bank is collecting to allow us to analyze the impact of the mix on checks and ACH files in productivity. The value of ACH files is not relevant as an input. Other inputs we would have to consider are “hours of personnel”, “technical sophistication of traditional facilities”, “technical sophistication of image facilities” (i.e. 1 if they have the latest technology, 0 if they do not) , “technical sophistication of ACH facilities”, as well as “availability of delivery” (i.e, 1 if they have easy access to train/plane, 0 if they do not).

[pic]

Quality:

The Collecting bank is exposed to Operational Risk, failing to process some checks and ACH files in the legal timeframe. There is a trade off between “how fast we process the checks” and “how many ACH files we process” (Productivity) and “how many checks and ACH files we failed to process on time” (Quality).

Therefore, “% of checks of size(i) processed in the legal timeframe” would be the output of the DEA Analysis. The size of the checks is also relevant from a quality perspective: big checks should be considered as premium clients of the Processing Center, and therefore should receive a better service in terms of quality. As size does not matters in ACH files, “% of ACH files processed in the legal timeframe” would be the output.

As for the inputs, we would use the same we used for Productivity.

[pic]

As for lost or illegible checks, they should be prevented to enter the system at the branch level (Step 1). The cashier is in charge of checking legibility if deposited in person. For checks deposited in ATMs, the branch should alert its customers if the branch sorting machines were unable to deal with the check, asking them to produce a legible document.

c) DEA Analysis of Step 3: Clearing House (Collection/Sorting/Delivery to the Paying Bank)

Productivity:

From the Clearing House perspective, productivity is proportionately related to the number of items (paper/image checks and ACH files) they process, as their fees are volume-based (both in the case of the Federal Reserve and Private Clearing Houses).

Therefore, “# of items processed/day” would be the output of a DEA Analysis. The size (value) of items would not matter from the Clearing House perspective because of the volume-based fee.

Inputs selected for the analysis were “# of entry points” the bank’s processing center is collecting items from (number of customers it collects from), “# of end points” (number of customers it delivers to), “total # of items”, “Mix of paper/image/ACH”, “hours of personnel”, “technical sophistication of traditional facilities”, “technical sophistication of image facilities” (i.e. 1 if they have the latest technology, 0 if they do not) , “technical sophistication of ACH facilities”, as well as “availability of delivery” (i.e, 1 if they have easy access to train/plane, 0 if they do not). Again, the major difference when compared with the inputs for the Collecting Bank’s Processing Center is the lack of relevance of check sizes.

[pic]

We have based our selection of productivity inputs and outputs in the DEA analysis the Fed conducted on its clearing facilities in 2002[17].

Quality:

The Clearing House is also exposed to Operational Risk, and there is a trade off between “how many items we process a day” (Productivity) and “how many items we failed to send to the right party” (Quality).

Therefore, “% of items sent to the right party” would be the output of the DEA Analysis. As for the inputs, we would use the same we used for Productivity excluding “availability of delivery”, which should not have influence in the outcome.

[pic]

d) DEA Analysis of Step 4: Paying Bank (Collection/Sorting/Check Availability of Funds/Debit Customer’s Account or Return Check to Collecting Bank)

Productivity:

From the Paying Bank’s perspective, productivity is a complex issue when dealing with paper and image checks. On one hand, the faster they clear the checks the more float they lose. On the other hand, Non-Sufficient-Funds fees (NSF) are the major source of a bank’s checking related revenues. Some 30% to 60% of the checking-related revenues of banks come from penalties charged to bounced checks. Therefore, Paying Banks have an incentive to clear “suspicious checks” soon, to increase the probability of the check to bounce (this is especially true for float-players, which count on clearing days to receive the funds they need) Suspicious checks are checks coming from clients that have a history of bounced checks. None of these issues appear with ACH files. For float calculation purposes, ACH transactions settle the day they are initiated. For sufficient fund checking purposes, the procedure is automatic and NSF penalties are not charged.

Therefore, “processing days/suspicious check” would be the output of a DEA Analysis. Size does not matter in this case either, as the NSF fee does not depend on the size of the check (around $20 per bounced check). “# of ACH files processed” would be the second output.

As for the inputs, productivity will heavily depend on the “total # of items”, “Mix of image/ACH or substitute/ACH” and the “% of suspicious checks”. Other inputs would be “hours of personnel” and “technical sophistication of facilities”. Additionally, we may need to consider the “# of points of entry” although it will vary on time (depending on which region customers are paying with checks to the Paying Bank customers)

[pic]

Because of the Check 21 Act Clearing Houses are obliged to work with substitute checks, no further distinction between paper and images is necessary. The bank will work either with Image and ACH or with Substitutes and ACH.

Quality:

The Paying Bank faces the trade off between “clear the bad checks fast” and “clear a lot of ACH files” (Productivity) to try to increase their NSF fees in checks and exceeding the legal timeframe to process the good ones to maximize float “how many checks we failed to process on time” (Quality)

Therefore, “% of checks(suspicious, not suspicious) processed in the legal timeframe” and “% of ACH files processed in the legal timeframe” would be the output of the DEA Analysis. In checks, the fact that the checks are “suspicious” or not is relevant for the Quality Analysis, as again, “suspicious checks” should be considered premium customers for the Paying Bank because of its superior profitability.

As for the inputs, we would use the same we used for Productivity.

[pic]

BIBLIOGRAPHY

• Fed to cut jobs as fewer use checks, 3rd August 2004, The Boston Globe

• Blue Book: Payment and Securities Settlement Systems in the European Union, European Central Bank

• Statistics on Payment Systems in the Group of Ten Countries, Bank of International Settlements

• The 2004 Federal Reserve Payments Study, Federal Reserve

• Humphrey, Willeson, Lindblom, Bergendahl, What does it cost to make a payment? Review of Network Economics, June 2003

• Uniform Commercial Code (U.C.C.) Cornell University (Law) Section 4

• Flatraaker, D. and P. Robinson (1995) Income, Costs, and Pricing in the Payment System, Norges Bank Economic Bulletin

• Gresvik, O. and G. Øwre (2002) Banks’ Costs and Income in the Payment System in 2001, Norges Bank Economic Bulletin, 73: 125-33.

• Hannan, T. (2002) Retail Fees of Depository Institutions, 1997-2001, Federal Reserve Bulletin

• Retail Fees of Depository Institutions 2003, Federal Reserve Bulletin,

• Gowrisankaran , FRBSF Economic Letter, Number 2002-27, September 20, 2002

• Gilbert and Wheelock , New evidence on the Fed’s productivity in providing Payment Services, Federal Reserve Bank of St. Louis, September 2002

• Gerdes, G. and J. Walton (2002) The Use of Checks and Other Noncash Payment Instruments in the United States, Federal Reserve Bulletin, 88: 360-74

• Wells, K. (1996) Are Checks Overused? Federal Reserve Bank of Minneapolis Quarterly Review, 20: 2-12

-----------------------

[1] Fed to cut jobs as fewer use checks, 3rd August 2004, The Boston Globe

[2] “Blue Book: Payment and Securities Settlement Systems in the European Union (ECB), Statistics on Payment Systems in the Group of Ten Countries (BIS), Federal Reserve

[3] The 2004 Federal Reserve Payments Study, Federal Reserve

[4] Blue Book: Payment and Securities Settlement Systems in the European Union (ECB), Statistics on Payment Systems in the Group of Ten Countries (BIS), Federal Reserve

[5] Verisign

[6] Statistics on Payment Systems in the Group of Ten Countries (BIS)

[7] Blue Book: Payment and Securities Settlement Systems in the European Union (ECB)

[8] What does it cost to make a payment? Humphrey, Willeson, Lindblom, Bergendahl, Review of Network Economics, June 2003

[9] The 2004 Federal Reserve Payments Study, Federal Reserve System

[10] What does it cost to make a payment?, Humphrey, Review of Network Economics Vol 2 Issue 2 June 2003

[11] Flatraaker, D. and P. Robinson (1995) “Income, Costs, and Pricing in the Payment System,” Norges Bank Economic Bulletin

[12] Hannan, T. (2002) “Retail Fees of Depository Institutions, 1997-2001,” Federal Reserve Bulletin, Retail Fees of Depository Institutions 2003

[13] Retail Fees of Depository Institutions, 1997-2001, Retail Fees of Depository Institutions 2003; Federal Reserve Bulletin

[14] FRBSF Economic Letter, Number 2002-27, September 20, 2002

[15] What does it cost to make a payment?, Humphrey, Review of Network Economics Vol 2 Issue 2 June 2003

[16] Flatraaker, D. and P. Robinson (1995) “Income, Costs, and Pricing in the Payment System,” Norges Bank Economic Bulletin

[17] New evidence on the Fed’s productivity in providing Payment Services, R. Gilbert and D. Wheelock, Federal Reserve bank of St. Louis, September 2002

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download