This evening I would like to share with you some current developments in the ways the Reserve Banks deliver payment and related services to banks such as yours. We have had to think strategically about service delivery this year, and I want to let you know what we are thinking and doing, and why.

I realize that some of you are familiar with how our electronic services work, and others are not. So, for some of you the particulars I will mention may be interesting. For all of you, I hope you can identify with the lessons we are learning about strategic focus and strategic choice.

One strategic choice that was difficult for us to make earlier this year had to do with what we call the "access channels" that banks use to do business with us electronically.

People in your banks can sit down at a PC and connect with us electronically to transfer funds, to look at account balances, and to get various kinds of information.

For about 15 years banks have used a combination of PC hardware, software, and telecommunications we call "Fedline" to access these electronic services.

This first generation of Fedline works pretty well, and has held up remarkably well over a lengthy period. However, it is based on an old operating system, called DOS, which some of you may remember, and for some time we have planned to replace it.

A couple of years ago, our plan was to replace this DOS Fedline with two new channels: Fedline for Windows, and Fedline for the Web. Our long-term direction was to provide our customers with what I will call "Internet-style" access to all of our payment and financial information services; maybe using the public Internet to gain access to a protected Federal Reserve "Extranet" environment. That would be Fedline for the Web.

However, there have been, and still are, serious security challenges around the actual transfer of value - the movement of money between accounts - using the Internet.

So, our plan was to bring forward a new version of our Fedline software that would run under the Windows operating system. Fedline for Windows would make use of special-purpose PCs and dial-up connections to the Federal Reserve's proprietary network, with much greater security than currently could be built into Fedline for the Web. We intended to put the value-transfer services, including funds transfer, securities transfer, and ACH electronic payments, on Fedline for Windows, and put informational services on Fedline for the Web. Meanwhile, we would keep working toward a more secure version of Web access for all of our services.

Earlier this year, we decided to change that plan. We decided not to go forward with Fedline for Windows, and to accelerate our work on Fedline for the Web. Three factors contributed to this decision.

For one thing, the development and testing of Fedline for Windows had taken longer than we had expected. It was a very good product, but it did not come along soon enough to have a lengthy useful life as the successor to DOS and the precursor to the Web.

Second, we would be requiring our customers to go through two conversions with us, or maybe even three conversions, if we went ahead with both channels.

And third, our growing pains with the DOS platform over the years, and with Windows in recent years, taught us that we needed to sharpen our strategic focus, and concentrate on where we really wanted to go strategically.

We decided that if we were trying to bring out one new channel for our customers and also trying to develop another channel, both might suffer. In particular, we were likely to delay the successful development and implementation of Fedline for the Web, which we regard as strategically important, because we believe it will give more flexibility and capabilities to the banks that use our services.

So, we are going to focus our attention and our best technical talent on Fedline for the Web, with the objective of having a solution for secure payment services, which banks can access via the Web, within the next two or three years. We will continue to support the DOS version of Fedline until we have that Web solution in place, and we will continue to move all of our informational services onto Fedline for the Web.

This was a difficult decision for us. We would like to leave the old DOS technology behind. We had worked hard on the Windows channel. We are taking a sizable risk by counting on a secure Web channel which does not now exist. However, we believe this new direction will help you, because we now will move faster to the Web platform that will serve you best.

And in facing up to this decision, we have learned something about strategic focus.

There is another big piece of our service delivery to you where we are making a strategic choice right now. That big piece is customer support for our services.

Historically, each of the 12 Reserve Banks has provided to its local customers all of the "help desk" support for Fedline, and for the services banks use via the Fedline channel. These are national services, running on centralized automation platforms, but teams of people at each Reserve Bank have helped to order the hardware, helped to install the software, answered all customer questions over the telephone, conducted training sessions, and done everything possible to help customers to resolve problems and work around disruptions.

We like to believe this local approach has served customers well, and has helped the Reserve Banks to stay close to their customers.

However, this approach is costly. We have pretty large groups of support people in all 12 Banks, and actually in some branch offices as well.

Recently we have been trying a more consolidated model for customer support. For Fedwire funds transfer and securities transfer, our Boston Reserve Bank now provides operational support to the banks of New England and the banks in five other Federal Reserve Districts. The Kansas City Reserve Bank serves the other half of the country.

For electronic ACH payments, the Atlanta and Minneapolis Reserve Banks provide operational support to banks across the country.

We have just implemented these consolidations during the past year, so we still are learning how best to do them, and still trying to assess the impact on our customers. Even as we do, we are considering how best to provide customer support for our new Fedline for the Web channel, which we are introducing now and therefore lends itself to some new approaches.

We know that consolidation of customer support from 12 Banks to 2 can reduce costs a lot. When we reduce costs, we reduce our prices to you. Funds, securities, and ACH prices all have been reduced in 2001 and 2002. Lower prices from us help banks to serve their customers more cost-effectively.

Our difficulty has been that we have seen the choice as being between the best customer service and the lowest cost. Each Reserve Bank takes pride in trying to serve its local banks well. From a Reserve Bank perspective, consolidation can look like a threat to the quality of customer support.

Meanwhile, payment services are very competitive. Our businesses must keep costs and prices down, and our customers always are trying to reduce their costs as they compete for corporate and consumer business. So, from the perspective of running the national funds transfer or ACH service, having support in 12 or more places is too costly.

Strategically, we have to bridge this "either-or" divide. We need to focus on lower cost and higher quality as one package to deliver to our customers. Easier said than done, but this seems to be the choice that makes the best sense.

If we do consolidation well, we ought to be able to implement best practices pretty quickly in two places, rather than waiting for people in all 12 Reserve Banks to do so. In fact, our strong local orientation sometimes has been an obstacle to implementing good ideas from elsewhere.

If we do consolidation well, we ought to be able to afford state-of-the-art technologies to support first-rate customer support. We might not be able to afford such technologies in 12 places, but we can in a couple of places that can serve all customers.

If we do consolidation well, we can build automatic backup into two support centers, so that one can carry on for the other, without interruption for our customers, if one of the centers has a disruption.

And if we do consolidation well, we will keep all Reserve Banks in close touch with their local banks. The Reserve Banks will not be providing all of the familiar operational support we have provided, but we still will be the Federal Reserve people closest to you. Our role will be to understand how you use our services; to understand what you need from the Reserve Banks to do your business better; to bring your needs, and new product ideas to meet your needs, into the continuous improvement of our national payment services; and, to make sure the Reserve Banks providing customer support for you are being held to high standards of quality and performance.

Our challenge has been to define the strategic direction. The best direction, I think, will be to implement high-quality, lower-cost support in ways that work so well that customers will have no concern about where the support centers are.

With this direction, you will continue to see our prices go down, you will receive excellent support, and you will be able to rely on the Federal Reserve as an effective service provider for the long run.

These have been tough issues for us, and I hope you find it a bit interesting to hear how we have addressed them. Thank you.