by Paul M. Connolly, First Vice President and Chief Operating Officer, Federal Reserve Bank of Boston
Massachusetts Association of Certified Public Accountants
April 28, 2005

This morning I would like to talk with you about important changes that are happening in the national payments system. Most people do not think much about the payments system, or even realize that there is such a thing. However, it is essential infrastructure for the financial system. It touches all of our lives. And it is changing in interesting ways.

I will be talking about what we call “retail payments”, meaning the mostly smaller-value payments made by consumers and businesses, but not including cash. Essentially, this means checks and electronic payments.

I would like to look with you at recent changes in U.S. retail payments; the significance of Check 21, a federal law that became effective last fall; and a few thoughts about what lies ahead. I will mention some considerations for financial professionals such as you. And you probably can think of issues and implications that I am not mentioning, so I will welcome your questions and comments at the end.

Three important ongoing changes I will discuss are the decline of the paper check; the growth of electronic payment choices; and the growth of home banking through personal computers.

We think the use of checks in the United States peaked sometime in the mid-1990’s, at a level a little above 50 billion annually. By 2000 a decline was occurring. Meanwhile, the use of the Automated Clearing House, or ACH, and the use of credit cards and debit cards were growing. If ACH is not a familiar term, that is the mechanism that processes direct deposit of pay, direct bill payment, and, as we will see, an increasing variety of other payments. In 2000, while these trends were evident, the volume of checks still exceeded the combined volume of these alternatives.

In this decade, though, change has continued, and accelerated. The Federal Reserve’s most recent Retail Payments Study, published in December of 2004, provides an update. By these estimates, the volume of checks written had fallen to about 37 billion by 2003. And the check share of retail payments had fallen below 50 percent. For the first time in our history, electronic payments had surpassed check payments. The ACH had continued to grow, to 9.1 billion payments; the mature credit card also had grown, to 19 billion; and the most recent entry, the debit card, had expanded most rapidly, to 15.6 billion.

Both offline debit card transactions, meaning transactions in which the customer signs a receipt; and online debits, where the customer uses a PIN number for security, had compound annual growth rates exceeding 20 percent from 2000 to 2003.

The ACH grew at a healthy 13 percent annual rate. The credit card grew by almost 7 percent. The check declined by 4.3 percent annually, with some reason to believe that the rate of decline was accelerating. Just over the three-year period from 2000 to 2003, the paper check declined from 57 percent to 45 percent of retail payments.

I think we can forecast that by the end of this decade, while we still will have a lot of checks, they will be an ever smaller factor in retail payments. Debit card transactions probably will have surpassed checks. Credit cards may be pretty close. And the ACH still has plenty of room for growth. In fact, we have seen growth in some new ACH options, from early 2002 through the end of 2004.

The “WEB” option enables consumers to authorize payments via the Internet, usually by paying a bill at a company’s Website, or by ordering something online and providing the bank account information needed for an electronic ACH payment. At the end of 2002 the Reserve Banks, which process roughly two thirds of the ACH payments in the U.S., processed just under 20 million WEB transactions monthly. Two years later, the number had grown to about 65 million per month. Most of these payments have replaced check payments.

The hottest product in ACH history has been the ARC, or Accounts Receivable Conversion. This is the option under which companies receiving consumer payments in lockbox operations take in checks through the mail, convert the checks into ACH transactions, and collect them electronically. At the outset of 2002 there were none. By the end of 2004, the Reserve Banks were processing over 90 million per month. The National Automated Clearing House Association, NACHA, has estimated that in 2005 there may be over 2 billion ARC transactions. In the short run, ARC transactions do not displace check writing, but they do take paper checks out of the bank collection system, and that is having an impact on correspondent banks and Reserve Banks. Over time, they probably will influence the check-writers to make some of these payments electronically in the first place.

Finally, the POP, or Point Of Presence, transaction refers to the conversions of checks into electronic payments at the cash register. In this model the merchant swipes the check to capture the MICR line with the account information, and hands the check back to the customer. This product has not been growing much. The largest retailers have shied away from it, because it requires some investment at every register; and it requires a customer’s signature, often with some confusion, and that slows down the checkout lines.

However, merchants are interested in reducing the number of checks they have to collect. Some merchants are working with NACHA on what they call a “back office version” of the ARC model. Instead of converting checks one by one at the registers, merchants could take all the checks received into the back office at the end of the day and convert them into ACH payments in one batch operation. If this idea comes into being, it could grow rapidly.

The biggest recent changes in consumer behavior that affect payments probably are the adoption of debit cards, as we have seen, and the adoption of online banking.

By some estimates, just 9 years ago only the truly committed users of personal computers were using them for home banking services – about 2.4 million households. That number had grown more than 12 times, to over 29 million households, by 2003. From this much larger base, the rate of growth will diminish, but steady growth will continue.

Moreover, the likelihood that users of home banking services will make more of their payments online is ever greater. Large billers are much more receptive to receiving consumer payments electronically than they were even 5 years ago. In fact, many of them encourage their customers to receive and pay their bills at the billers’ websites. Thanks in large measure to the ACH, which provides low-cost, reliable electronic connectivity for payments for nearly all financial institutions in America, banks now can complete most of a consumer’s electronically initiated payments electronically. And banks are promoting their home banking services more aggressively.

Home banking will be another growing force for replacement of paper checks with electronic payments. That trend is good for the country. A mostly electronic payments system will be less costly and more reliable.

Meanwhile, we still have those 37 billion checks. Each year we will have fewer of them, but we still will have billions for many years.

Checks are costlier to process and collect than electronic payments. They take longer, because the check collection system depends on the physical delivery of the paper to the check-writer’s bank. And sometimes bounced checks take a long time to come back to the parties who accepted them, which brings risk of loss and opportunity for fraud.

As of last fall, the country has a new means to alleviate these problems in the check collection system. It is the new Check 21 law.

This new law responds to a long-standing interest among banking industry leaders to find a way to collect checks electronically. This interest in increased efficiency had to be balanced with existing check law, the Uniform Commercial Code, and the interests of consumers and businesses.

The case for legal change received new impetus from the awful events of 9-11. You may remember that for most of that week in September, 2001, the nation’s airports were closed. Among other things, this meant that many checks could not be collected, because check collection depends so much on transporting paper from one city to another. This vulnerability caught the attention of members of Congress, who recognized that the country needed an alternative that would reduce this exposure for a payment mechanism used by so many of us.

It took two years to bring together banks of all sizes, consumer groups, business interests, attorneys, and others to craft what became the Check Clearing For The 21st Century Act, or Check 21. In October of 2003 it passed both Houses of Congress, with no dissenting votes. President Bush signed it into law, with the effective date of October 28, 2004.

Check 21 enables, but does not require banks to collect checks by means of transmitting electronic images, or pictures, of them. To supplement the use of images, the law creates a new instrument, called the substitute check.

A substitute check is a paper reproduction of an original check. This reproduction is made from the digital image of the check. So, for instance, a bank in San Francisco with a check drawn on a bank in Boston, can capture an image of the check and deliver it electronically to the Boston bank, or to a third party that in turn will present it to the Boston bank. That third party could be the Federal Reserve Bank of Boston, or a correspondent bank. Note that this image transmission can be done in seconds, whereas the physical delivery of the paper check would have taken many hours, probably overnight.

If the bank in Boston agrees to accept the image as the legal presentment of the check, that will happen. If the bank wants presentment in paper form, it must accept a properly prepared substitute check.

The substitute check includes all of the information from the front and the back of the original check. It also has, along the bottom edge, those funny-looking numbers and symbols needed to process paper checks through high-speed sorters. Those numbers must be printed with the magnetic ink required by check processing equipment. And the substitute check also bears specific language to explain its legal standing.

When prepared in accordance with standards, the substitute check is the legal equivalent of the original check. A bank, and the bank’s customer who wrote the check, can use this substitute check for any and all purposes for which they would have used the original.

So, the essence of Check 21 is to provide an avenue banks can choose to use to reduce the extra handlings and the lost time required by traditional paper check collection. The substitute check provides a safety valve, to ensure that images and paper can coexist.

Some of the largest banks in the country were preparing to make use of the law’s provisions well before last October. They have begun to collect some checks via image transmission, usually small volumes of larger-value checks for which faster collection has some significant value. Many other banks are in various stages of preparation and experimentation.

One factor that slows progress is software. Banks and their suppliers need new or modified software to be able to send and receive images in the format required for Check 21. Some banks also want to be able to print substitute checks, and that is not simple to do. In addition, new controls are needed to ensure that when a check is collected in image form, the original paper check is not also sent forward for collection.

These changes take time, but we see change happening. Right now about 75 banks across the country are using Check 21 services offered by the Reserve Banks, and many more are in our testing queue.

The Reserve Banks are collecting over 400,000 checks each night via Check 21. This is less than one percent of our nightly volume. However, the average value of these checks is over $13,000, much higher than the average value of all checks. Again, this indicates that so far banks are using Check 21 to accelerate the collection of larger-value payments, most of which are checks issued by businesses rather than consumers.

In thinking about Check 21, banks will be sensitive to the impact on consumers. At a minimum, as time passes consumers are likely to see a few substitute checks showing up in their monthly statements, and this could cause some confusion. Also, over time some of a consumer’s checks may be collected a day faster through Check 21. If the consumer is counting on a longer period before the check “clears”, he may be surprised by a bounced check or an overdraft charge.

An even larger change to consider is whether to promote to consumers the use of image statements in place of the return of cancelled checks. As the use of images to move checks between banks comes into greater use, and as banks enhance their internal systems to be able to work with images in addition to paper, more of them may become more active in promoting image statements, and home banking, to give consumers the information they need without mailing those cancelled checks.

Many banks will be weighing the opportunities, the risks, the costs, and the benefits of Check 21 during the next year or so. Meanwhile, Check 21 will take hold and grow.

By 2007 or so I believe we will see more broad-based adoption of Check 21, and steady growth in the use of images to collect checks. Then, at some time, maybe late in this decade, we will reach a “tipping point” where the majority of checks are collected electronically. When that happens, other changes will follow.

To take a possible example, when many of the checks that used to require air transportation no longer require it, that transportation probably will be too expensive for banks with fewer checks riding on it. So, the banks will have incentive to use Check 21 even more; or to discourage the use of checks; or to ask check-writers to pay more to use checks.

Every action has a reaction, so consumers and businesses will make choices in response to such changes. Most of the changes we can imagine will accelerate the use of electronic alternatives to the check.

Another force for change, and for the growth of Check 21, is the transformation taking place in the Federal Reserve’s national infrastructure for check collection.

The Reserve Banks collect about half of the checks in the U.S. that we call “interbank” checks, meaning that one bank has to get the check presented to another bank for payment. As the use of checks declines, our volumes decline. As checks written get converted into electronic payments, such as in the “ARC” model, our volumes decline further. We favor these trends. We have been advocates of a more electronic payments system. However, the transition is very challenging to manage.

In response to volume declines, we are consolidating our check offices around the country. We came into 2003 with 45 check processing locations. By early 2006 we will have 23. As volume continues to decline, we will consolidate further.

Here in New England, we closed a small office we had in Maine in 1997, in anticipation of the changes we see today. In February of next year, we will consolidate the check processing we do in downtown Boston into our facility in Windsor Locks, Connecticut. The other Reserve Banks are doing similar consolidations.

Please note that we will continue to offer check collection services to every bank in the country. We will just so do from fewer processing locations.

Locally, we already are seeing that our announced consolidation into Windsor Locks is spurring a lot of interest in Check 21 among New England banks. For many of them, having their checks travel to and from Windsor Locks instead of Boston will mean some disruption in their normal back office processing schedules, and some extra travel time for the checks. We will work with the banks to minimize the impact, but there has to be some impact.

Check 21 will help. If banks send check images electronically instead of transporting paper, they can offset the greater geographic distance. I think the Federal Reserve’s infrastructure changes in response to declining volumes will help Check 21 to grow.

These are exciting times in retail payments. The pace of change is accelerating. Consumers and businesses have new choices for making their payments. More choices keep emerging. Look, for instance, at PayPal, a new online payment option. Also look at the “fob” credit cards, the key-ring-size cards you can just wave at a reader to charge an expense in places that never accepted credit cards previously, such as coffee shops and fast-food chains.

With change comes some uncertainty. Here are a few questions about the future.

Is Check 21 a mechanism that will last for decades? Or is it an interim step on the way to an overwhelming electronic payments system? Right now, nobody really knows. Check 21 is needed now. We have those 37 billion checks, and even when 50 percent of them are gone, we will have close to 20 billion. Check 21 provides for electronic collection while respecting consumer choice. If the pace of change continues to accelerate, Check 21 may not have a big role, say, 10 years from now, if America does not have huge volumes of checks by then. Today, we can only speculate.

Will check image capture and receipt extend beyond the banks, to businesses and consumers? For businesses, definitely. We will see businesses that take in checks scanning them to capture images, and then transmitting the images to their banks for collection. For consumers, we will see the equivalent at ATMs. When a consumer deposits a check at an ATM, some ATMs will capture an image and transmit it to the consumer’s bank. In fact, this concept was one of the seminal ideas from the banking industry that gave rise to Check 21. Whether a consumer will be able to do the same thing from her home computer, I do not know. However, it is conceivable.

Will “ARC” and Check 21 both grow as lockbox solutions? The “ARC” model applies only to check payments from consumers. For business-to-business payments made by check, companies and their banks may decide to use Check 21 to collect those checks electronically.

Beyond that, some people have speculated that Check 21, and image transmission, once they get going, will replace the ARC model. Personally, I doubt it. The ARC model seems to work well. It gives companies a lower-cost way to collect check payments than they had previously. And it has been accepted by nearly all affected consumers.

Just a few years ago it would have been difficult to think this thought seriously. However, it seems clear now that we are headed for a retail payments system in which the paper check will be a minor element. Electronic payments already are in the majority. More electronic choices are coming forward. And there seems to be a demographic difference, between older and younger people, when it comes to check-writing. For people under 30 or so, college students and young adults in the workforce, writing checks seems to be what they do as a last resort when they cannot pay electronically. To them, the check for payments may be like a vinyl record for music: something their parents used to use. Many older people use electronic payments, too, of course. Still, generational change seems likely to spur the growth of electronics.

I would like just to touch on a few considerations for financial professionals. First, for your business clients, they will start to see some of the checks they issue come back to them as substitute checks. They will look different, but legally there will be no difference. You can assure them that all of their rights are protected. If you have clients with a lot of checks to collect, you can advise them to ask their banks about using Check 21 to collect the funds faster.

For any individual clients you may have, and for yourselves, you should know that the Check 21 law provides extra protection for consumers. Of course consumers have all the same rights and protections with substitute checks that they have with their original checks. In addition, they have what is called the right to “expedited recredit”. If a consumer receives a substitute check instead of her cancelled original check, and believes that the substitute check was charged to her incorrectly, and she has lost money as a result, and she needs the original check to show that the substitute somehow is in error, then this special protection applies.

When she makes such a claim to her bank, the bank of course must look into it. If the bank cannot resolve the issue within 10 business days, the bank must refund the amount of the loss up to the value of the substitute check, or $2,500, whichever is less. If the question still is not resolved after 45 calendar days, the bank must refund any remaining amount of the substitute check. If the bank determines later that her claim was not valid, the bank can reverse these refunds. However, this special protection favors the consumer while a claim is under investigation.

One concern about Check 21 has been the possibility of making some forms of check fraud harder to prevent or combat. Recipients of images or of substitute checks will not have the original pieces of paper to examine, for weight, or texture, or pen pressure, or fingerprints. While this is true, this kind of examination does not seem to be part of most fraud investigation or detection.

The best ways to deter check fraud are for the party accepting a check to verify the identity of the check-writer, and know how to reach him if the check bounces; for the bank accepting a check deposit to know its customer; and for the check collection and return processes to work swiftly.

Check 21 will help to speed check collection, and speed the return of bounced checks. It is likely to deter far more losses and fraud than any it may facilitate.

As I mentioned, the “ARC” model for conversion of checks into electronic ACH payments does not apply to corporate checks. Positive pay systems and other services banks provide for business customers could be impaired if the check payments came back as ACH debits. Also, some banks reject all ACH debits against corporate accounts, to protect business customers against fraudulent transactions.

However, the phenomenal success of the ARC model with consumer payments has prompted some banks and businesses to look at how it might be used to collect some corporate checks, without impairing established services and safeguards. These explorations may or may not bear fruit, but you might want to stay tuned.

And what will happen to float? With the relatively low interest rates of the past several years, we have not heard as much about check float as we used to hear. Still, some businesses and consumers count on it. With Check 21, float in the bank collection system will be reduced. On a very small scale this already is happening with the larger-value corporate checks I mentioned. Of course there still is float while “the check is in the mail”, and when the recipient does not deposit the check for collection immediately. So, float will not be eliminated, but it will be diminished.

Now, just a few closing thoughts I would like to share. The decline of the check illustrates that big changes often take longer to happen than we expect, and then happen faster than we expect. Many people were humbled during the past 40 years by their predictions of a “checkless society”. The use of checks in the U.S. grew for many more years than some people had expected. Now, the transition to electronics is happening, and the pace is accelerating.

Second, new and old payment choices co-exist for very long periods. The check as we know it in America is about 150 years old. The credit card is over 50 years old. The ACH is 30 years old. Until recently they all grew together. And we expect all of them, and the debit card, and numerous other choices, to co-exist for a long time to come.

However, dramatic change is happening now. I hope our time together this morning helps you to be more aware of how the everyday world of retail payments already has changed a great deal, and of the substantial changes yet ahead.

Finally, I think we will see similar changes in the use of cash. We rarely think about currency and coin, but we make about four times as many payments with cash as we do with checks and electronics combined. Many are very small value transactions, such as for the newspaper, and for coffee. However, it costs society a lot to use cash. Think about bank vaults, and security in stores, and armored trucks. It costs a lot just to handle and count all of that paper currency. And electronic options are coming along to displace some use of cash. For instance, the debit card in the supermarket check-out line has replaced many cash as well as check transactions.

We always will have cash, but the emerging “less check” society probably will be a “less cash” society as well.

I hope this look at the changing payments system has been an interesting change of pace for you. As I said at the outset, I welcome your questions and comments. Thanks very much.