美國運通 (AXP) 2004 Q3 法說會逐字稿

Good afternoon. My name is Miles, and I will be your conference facilitator today. At this time I would like to welcome everyone to the American Express third quarter earnings conference call. All lines have been placed on mute to prevent any background noise. After the speakers' remarks there will be a question-and-answer period. Instructions will be given at that time as to how to ask a question. Thank you. Mr. Stovall you may begin your conference.

Okay. Thank you Miles and welcome to everyone. Appreciate all of you joining us for today's discussion.

As usual I just need to do a little housekeeping before we get started and remind you that the discussion today contains certain forward-looking statements about the Company's future financial performance and business prospects which are subject to risks and uncertainties and speak only as of today. The words believe, expect, anticipate, optimistic, intend, plan, aim, will, should, could, likely, and similar expressions are intended to identify forward-looking statements.

Factors that could cause actual results to differ materially from these forward-looking statements including the company's financial and other goals are set forth within today's earnings press release which was filed in an 8-K report and in the Company's 2003 10-K report already on file with the Securities and Exchange Commission.

In the third quarter 2004 earnings release and supplement which are now posted on our Web site at ir.americanexpress.com, and on file with the SEC in an 8-K report, we have provided information that compares and reconciles the managed basis, financial measures to be discussed today with the TRS GAAP financial information as well as AEFA's GAAP and net revenues and explains why these presentations are useful to management and to investors. We urge to you review that information in conjunction with today's discussion.

Gary Crittenden, Executive Vice President and Chief Financial Officer of American Express, will provide some introductory remarks highlighting the key points related to today's announcement. Once he completes his remarks we will turn to the moderator who will announce your opportunity to get into the queue for the Q&A period. Up until then no one has actually registered to ask questions.

While we will attempt to respond to as many of your questions as possible before we end the call, we do have a limited amount of time, and based on this we ask that you limit yourself to one question as a time during the Q&A. With that, let me turn the discussion over to Gary.

Thank you, Ron, and welcome everyone and thanks for joining with us today.

As I start, I wanted to thank those of you who gave your time to participate in our financial reporting survey interviews which took place in August and September. These surveys were part of a broader effort that is focused on identifying disclosure, presentation, and process changes to enhance the transparency and the value of the Company's reporting activities.

While survey responses and other benchmarking efforts indicate that overall you all think that we have better than industry average reporting activity at American Express, there are some good suggestions to further improve the transparency of our disclosures. Most of the related changes to our documents will be implemented in the fourth quarter of this year and in the first quarter of next year.

However, one suggestion that came out of the survey was to keep my opening comments on this call short and focused, so I plan to be somewhat briefer than past calls and I will allow our earnings release and supplement to provide some of the data that I have included historically, and then allow the questions and answers to focus our discussion on the items that are of greatest interest to you.

Moving on to our quarterly results. As you have already seen our third quarter results exceeded all of our long-term targets. Revenues grew 12% versus last year, net income increased 14%, our diluted earnings per share of 69 cents marked a 17% increase, and return on equity was 21%.

In addition, our shareholders have been rewarded by strong business performance. Year-to-date we have returned 89% of the total capital that we've generated, including a 20% increase in the quarterly dividend to 12 cents a share as announced on September 27th.

This total return to capital exceeds our long-term target of 65%. And between this dividend and the dividend we did last year, we've had a total increase in our dividend of about 45%.

Strong revenue growth was driven by a 16% increase in card member spending, 8% growth in worldwide lending balances on a managed basis, a 23% increase in travel sales, and a 13% increase in assets owned, managed, and administered at AEFA.

On the expense side, consolidated expenses were up 11%, continuing to reflect higher business building expenditures and human resources costs. Our decision to expense stock options beginning in the first quarter of last year has driven the higher compensation related costs, but the underlying human resource expenses and related employee count, continue to be well controlled.

We're on track to deliver the $1 billion of re-engineering benefits targeted for this year.

I will say a few words at the end of this discussion about our agreement to extend our partnership with Delta, but let me first review some of the third quarter business unit specific details.

At TRS, managed net revenues increased 11% and net income rose 20%. The benefits of our investment spending continue to be evident in our card-related metrics which are strong both on an absolute basis and versus the competition. Worldwide billed business continued to grow at a robust pace on continued strength in all of our segments.

In the U.S., for example, consumer spending grew 14%, small business spending rose 18%, and corporate services volume improved by 10%. Outside the U.S., reported billed business was up 21% which equated to 15% growth on a foreign exchange adjusted basis, reflecting a 12% growth in proprietary consumer and small business volume, a 14% increase in corporate services spending, and an increase in network partner related volumes of 27%.

Worldwide cards in force grew 7%, and average spending per basic card in force increased 11%.

The discount rate increased 1 basis point versus the second quarter of 2004, and decreased 3 basis points from the third quarter of last year as the mix of spending continued to evolve towards everyday spend categories.

Worldwide lending balance growth versus last year was solid, rising 8% to $45.6 billion, despite an industry environment that continues to challenge organic receivables growth. While airline ticket prices have declined, traveling revenues rose 20% on an improved sales environment, and the benefit of the Rosenbluth acquisition.

On the expense side, marketing promotion rewards and card member services expenses increased 28% as rewards costs rose and we continue to focus on other business building investments. Today, managed provisions for losses declined 3%, as charge card and lending credit quality as well as our reserve coverage ratios remained very strong.

Other provisions increased significantly primarily due to a charge of $115 million which was incurred as a result of the reconciliation of securitization-related lending receivable balances that had accumulated over a five-year period. This charge was net of $32 million worth of reserves previously provided.

An internal operational review of our securitization activity identified a computational error that caused a systematic understatement in the basis of our securitized receivables.

This error was immaterial in terms of both of its amount and its effect on the trend line in earnings and it was spread over 22 quarters. It was not discovered until this quarter.

This charge cumulatively adjusts the impact of these securitizations on certain retained lending receivables. Since identifying the error, we have performed a comprehensive review and have revised our procedures accordingly.

In addition, a benefit of $60 million was realized reflecting a reduction in merchant-related reserves. This adjustment reflects modifications to the Company's agreement with certain merchants to mitigate our loss exposure where the Company has paid the merchant but the merchant has not yet delivered the goods or services, so the Company is exposed to the merchant's credit risk.

In addition, it reflects our ongoing favorable credit experience with merchants over the last few years.

Human resources and other operating expenses rose 8% due to greater management incentive and employee benefit costs, merit increases and the impact of employees added through the Rosenbluth acquisition.

The effective tax rate of 31% decreased from 32% in the second quarter of this year and the third quarter of last year. This effective tax rate reduction is primarily the result of both one-time and ongoing benefits related to the restructuring of certain foreign operations as well as adjustments to foreign tax expense to reflect the results of completed tax returns.

At AEFA, pretax income increased 14% on growth in managed net revenues of 20%, however, net income declined 6%.

The Threadneedle acquisition contributed approximately 7% to revenue growth, and modestly to net income growth during the quarter. As the Threadneedle acquisition was completed on September 30, 2003, this will be the last quarterly disclosure in which we will be reporting its impact.

Assets owned, managed, and administered rose 13% on market appreciation and asset inflows. Total cash sales increased 11% on continued strength in mutual fund sales and the benefit of the Threadneedle institutional activities.

Management and distribution fees rose 21% on a 34% increase in management fees, and a 7% growth in distribution fees. Net investment income rose 5% due to net investment gains of $11 million this year, versus losses of $13 million in the third quarter of last year. Average invested assets rose 2% while the portfolio yield was essential flat.

Other revenues rose on strong property, casualty and higher life insurance revenue. The provision for losses and benefits decreased 3%, primarily due to a reduction in annuity product provisions resulting from lower crediting rates and the effect of depreciation in the S&P 500 on equity-indexed annuities partially offset by a higher average in force level.

Human resources and other operating expense increased 22% on the effect of the Threadneedle acquisition, higher field force compensation-related costs, higher advertising costs, and various security, industry, regulatory and legal matters.

We also recorded a net benefit of $24 million, which is about $15 million after tax, resulting from deferred acquisition costs-related adjustments arising from AEFAs annual third quarter DAC review. This compared to a $2 million net benefit last year.

The advisor base rose 3% versus last year and 1%, or 128 advisors from June of this year. The home office employee count continued to be well controlled as the average number of employees was flat excluding Threadneedle.

The effective tax rate was 28% this year, versus 12% last year, which included a $29 million favorable tax adjustment.

At American Express Bank, earnings grew 18% on 3% growth in net revenues and a lower provision. The bank's results reflected the positive impact of the growth within the financial institutions group and private banking area, which were offset by reductions in loan and other activity within corporate banking, and the personal financial services lending business.

Private banking client holdings and loans increased 9% and 3% respectively. Loans within the financial institutions group grew 16%, personal financial service loans declined 5%, and corporate banking loans continued to decrease and now represent 1% of the total loan portfolio or approximately $100 million.

Results also benefited from a provision decline of 46% versus last year, due to an improvement in bankruptcy-related write offs in Hong Kong and the reduced activity within the PFS lending portfolio.

Let me now address briefly our decision to extend our multi-faceted partnership with Delta Airlines and support their financing needs.

The agreement referred to in the press release extended our co-brand, MR and merchant relationships with Delta into the next decade and will position us to continue to grow a portfolio which, while less than 10% of the Company's worldwide billed business and less than 15% of our global lending receivables, represents a very attractive, high spending, and loyal card member base with excellent credit quality.

As part of these agreements, we will be prepaying $500 million for the future purchase of Delta SkyMiles rewards points, and we will be providing $100 million loan to Delta as part of a new credit facility that is current being negotiated with other lenders.

The prepayment will have a three-year term and both the prepayment and the loan will be fully collateralized by a pool of assets that are subject to certain conditions, including the successful completion of negotiations with the pilots union, restructuring of debt, and the finalization of a new credit facility.

As you know, Delta has publicly indicated that unless these conditions are met, they will file for Chapter 11 bankruptcy protection. If that happens, we will still provide the same level of support in the form of debtor in possession financing.

We believe the risks involved with this agreement are manageable. We have extensive experience in working with the airline industry, and have not had any significant losses from airlines that are experiencing financial challenges. We will have an appropriate level of collateral and we're operating from a position of financial and business strength.

Most importantly, pursuing this agreement with Delta presented a prime opportunity with significant mutual benefits upon which we could capitalize given our current performance and position in the marketplace. The extension of successful partnership relationships has been a goal of ours as indicated by our recent ten-year agreement that we have signed with Costco.

Separately, as you saw last Friday, we also announced an agreement to sell our small business equipment leasing unit, American Express Business Financial Corporation, with a portfolio of approximately $1.5 billion. We do not expect the gain on the sale of this business to have a material impact on the fourth quarter results because of unrelated newly anticipated costs associated with our global re-engineering activities.

In addition to that, last Thursday we were very pleased to announce the addition of Jet Blue to our Membership Rewards partner roster.

With that, let me now conclude with a few final comments.

We view this as another excellent quarter that illustrates the benefits of the business momentum achieved through our investments over the last few years. We delivered record results during the quarter and met or exceeded all three of our long-term financial targets while increasing business building expenditures and maintaining substantial balance sheet strength.

Although year-over-year business comparisons will again be more difficult in the fourth quarter of this year, and varying opinions exist about the future strength of the macroeconomic environment and equity market, we are in a very strong competitive position. We have a number of excellent growth opportunities on the horizon.

To help ensure that we have the resources to take full advantage of these opportunities, we plan to intensify our focus on re-engineering activities for the remainder of this year and into next year. Our recent business success make us confident that we are positioned to leverage these growth opportunities.

Thanks very much for listening. And I think we're now ready to take your questions.

Thank you, sir. At this time I would like to inform everyone in order to ask a question please press star then the number one on your telephone keypad. If you pressed star one prior to this point your signal did not register and you will need to press star one again. Please hold one moment for your first question. Your first question comes from the line of Chris Brendler with Legg Mason.

Hi, Gary. Good afternoon. Can you give us a little commentary, if you would, on the, I guess a little bit of a slowdown in the U.S. lending, or the worldwide lending, we don't break out U.S. separately. Maybe if you can give us some color on the U.S. market as well as what you're experiencing on the teaser front and maybe also if you think that the majority of the competition and competitive pressure in the credit card business is coming from mortgage financings sources like home equity lending and refinances or some of the big aggressive large banks who are mailing some pretty significantly attractive teaser offers?

Chris, I think primarily if you go back and kind of look at the billed business last year coming into this year, if you look at first part of last year, total card business was up 10% in the first quarter last year, 10% in the second quarter, and then up 15% and 17%, so our comparisons year-over-year have gotten, you know, a little stiffer as we've come through the year, so we were up 14% in billed business in the third quarter on top of being up 15% in the third quarter of last year.

So frankly we feel like the business momentum has continued at more or less the same level that we had seen earlier in the year, was just a little bit more difficult environment that we were competing against.

On the, in terms of teaser offers this is, as you know, an extremely competitive environment and the rates that you see in the marketplace change all the time. What you'll see is you'll see offerings for three or four weeks or a month that will have a certain characterization associated with them and literally three or four weeks later those will change.

We don't think that there has been any particular change in that kind of competitive dynamic over the last few months. What you say is true, the people who happen to be mailing the most today are a little different than the people who were mailing the most a year ago.

It happens to be the banks that are doing it right now, but there's nothing fundamental about what's happening in the marketplace that I think has changed our feeling for how we feel.

I do think it's interesting, if you look overall at the revolving credit market for the industry overall it's been in low single digits and our growth rate was 8% in this quarter. So overall, we feel like we continue to have very good performance, cards in force up 7%, billed business strong, the underlying dynamics of the business appear to be very healthy.

Great. Thanks.

Miles? Miles?

We lost our operator.

Who is that speaking?

That was Ed Groshans.

Were you next for the question?

I do have a question. I didn't know I was next. I'm sorry. I didn't even know I was live. Okay.

Go ahead.

My question's focused around AEFA. It seems like we had sequential declines in all the top-line revenue items there. In addition, sequentially it looks like we had some asset flows in the other owned accounts and the other separate accounts. I was wondering if you could just address those and get some color around that?

You know, I don't know if we would characterize the environment as quite as difficult as you might have just now, Ed. Let me pull my, some of my sequential charts here for just a second. I'm going to ask Ron to pull some of these sequential charts out so I can just take a look at it. But overall from quarter-to-quarter if you kind of compare how our overall asset growth rate stacked up versus most of the competition, we feel like we did reasonably well up 13%.

We have continued to have some negative outflows in our long-term equity mutual funds over the course of the last year, that kind of continued unabated, but broadly speaking, in spite of the fact that the equity market has not been particularly robust from an insurance perspective, for most of the underlying businesses, we feel like the performance has been pretty much in line with the industry. I'm just tracing across the exact numbers here.

Management and distribution fees obviously is a function of, you know, what our total assets are, and those were down from 752 to $733 million in this quarter. We have, you know, if you look at the mix of things, of the institutional side of our business with the, what Threadneedle has brought, has changed somewhat the composition of what those distribution fees have been around, you know, more institutional in their nature, but generally I mean there's, we don't think or wouldn't characterize it as being weaker than it's been over the last couple of quarters.

Thank you.

Your next question comes from the line of Michael Cohen with Susquehanna Financial.

In addition, sequentially it looks like we had some asset flows in the other owned accounts and the other separate accounts. I was wondering if you just address those and give some color around that?

You know, I don't know if we would characterize the environment as quite as difficult as you might have just now, Ed. Let me pull my, some of my sequential charts here for just a second. I'm going to ask Ron to pull some of these sequential charts out so I can just take a look at it.

Overall, from quarter-to-quarter, if you kind of compare how our overall asset growth rate stacked up versus most of the competition, we feel like we did reasonably well, up 13%. We have continued to have some negative outflows in our long-term equity mutual funds over the course of the last year, that kind of continued unabated, but broadly speaking, in spite of the fact that the equity market has not been particularly robust, from the insurance perspective, for most of the underlying businesses, we feel like the performance has been pretty much in line with the industry. I'm just tracing across the exact numbers here.

Management and distribution fees obviously is a function of, you know, what our total assets are, and those were down from 752 to $733 million in this quarter. We have, if you look at the mix of things, on the institutional side of our business with what Threadneedle has brought, has changed somewhat the composition of what those distribution fees have been around, be they're more institutional in their nature. But generally, I mean we don't think or wouldn't characterize it as being weaker than it's been over the last couple of quarters.

Alright. Thank you.

Your next question comes from the line of Michael Cohen with Susquehanna Financial.

Hi guys. Gary, can you provide a little more context on the computational error surrounding securitization? I mean I don't want to make too much of this but I don't necessarily want to make too little of it, either.

No, it's a very fair question. What essentially happened is we installed a new securitization income system, a system that would track our income from securitization back in the late 90s, and when that system was installed, there was an error in the way the program was written that resulted in our systematically understating the basis in the receivables that we securitized. And it was, in any one individual quarter, a relatively insignificant amount, but it did happen in each quarter and it accumulated over time.

Because it was regular and pretty systematic, it was difficult to uncover until we were going through a normal operational review, which we completed in the third quarter of this year. And when we did that, we realized that this basis had been understated, and as a result of that, we had overestimated income during this time period and made the decision, obviously, to take the charge that we took in this quarter.

So we took the charge of $115 million, we had already reserved for reconciliations. We carry a normal reconciliations reserve on our balance sheet, obviously. We took a normal reconciliations reserve against it as well so in total it was $147 million.

And then we completed a very comprehensive review around all of these systems obviously, to ensure that there was not another issue out there that we had missed in some way. But that's basically what it was.

And just to clarify, I mean, I know you just sort of said this, but the reason it flowed through the reserve line item, is this sort of reconciliation reserve as opposed to being a [contra] to revenue somewhere?

Yeah, you're exactly right. Whenever we have a balance sheet reconciliation item we always take that through this other provision account. That's where we set up our reconciliation reserve and we always take impacts like this through that same line item.

And we have these from time to time over the years but they're obviously much smaller than this one. This was a particularly large one because it had accumulated over such a long period of time.

Great. And if I may ask one other question, you know, I noticed, I guess, on a per unit basis in terms of spending per card in force you went down slightly in the quarter. Any color commentary on that?

Well, if I look at spending for card in force, I'm just trying to think, spending per basic card in force on page seven of my supplement here is up 9.

I'm talking about sequentially quarterly, on a sequential quarter basis.

You know, really, no. It's so dependent on what the mix is of cards in any one quarter that it really, you really can't say. It depends on a lot on what our acquisition activities have been during the course of that quarter. The thing I would really focus on is spending per basic card in force. And that, on top of a very strong growth last year was up again 9% this quarter.

Okay. Thank you.

You bet.

Your next question, sir, comes from the line of Craig Maurer with Fulcrum.

Good afternoon. Two questions. I was wondering if you can discuss the overall direction of the discount rate as it's been pretty flat for the past three quarters after you had been originally been discussing it coming down over time? And in addition if the recent regulatory issues surrounding the insurance industry could have any effect on AEFA's business? Thank you.

Sure. In terms of the discount rate, we continue to think it's going to follow more or less the same trend that it has historically, this kind of 3 or so basis point decline per year is driven very much by the mix and so in any individual quarter you might see some flattening like you did in this quarter.

In fact, if you actually disaggregated the discount rate between this quarter and last quarter, last quarter tended to round down, this quarter tended to round up, and so they ended up being basically the same number, but it really is just a function of the growth and performance in individual industry.

But I wouldn't read into this relative flatness over the last couple of quarters that that's likely to continue forward. We think the expansion into everyday categories is likely to continue this gradual downward trend.

In terms of the insurance industry, obviously there's a number of different issues that are happening there today. As far as we know those are generally not the kinds of businesses that have any involvement with AEFA so we manufacture almost all of our own insurance products, they've sold through our own captive distribution. We don't have much of a third-party business where we sell outside of our network.

In terms of the property and casualty business we typically sell that only directly to our own card members or through Costco, and so, you know, if you just take kind of the block of issues that currently surround the industry, we don't appear to be generally involved in those kind of activities.

Okay. Thank you.

Your next question comes from the line of David Hochstim with Bear Stearns Company.

I had one clarification on a question. The question is basic cards outside the U.S. didn't grow this quarter and we've had kind of a gradual slowing. I wondered what's going on there, if there had been a shift in marketing spending?

And then clarification was just on the Delta financing. Were you saying that it's only partially secured by assets? Or related to that, could we expect some expense saves if you're buying $500 million worth of miles up-front, that's eight or nine billion miles or something that maybe you get a better deal just as you offer cheaper to your customers involved?

Let's see. In terms of the, outside the U.S. card growth, we have really renewed our focus outside the U.S., David, on high-value card members. We have really put a focus on bringing in the highest spending customers outside the U.S. as we possibly can.

And that shift in marketing direction has required some digestion, and so you're right, you are seeing some slowing in the basic card member account outside the U.S., but it really is part of what we're trying to do strategically and directionally it makes a lot of sense for us.

We'll obviously continue to fine-tune that strategy here over the next few quarters, but what we're trying to do, obviously we'd like to have the card number be up higher than that, but it does represent a bit of a reorientation in terms of how we're spending.

So the growth in total cards is really just getting supplemental cards to the highest spenders you already have?

That's correct. That's a correct interpretation of that number for international.

On the Delta side, no, it is the, the account is, the total credit line is completely secured, it's not just partially secured. We have what we believe to be very sound collateral that would take into account this entire obligation that they have towards us.

The 100 or the 600?

The entire 600 million is secured.

As far as the purchase up-front goes, frankly, we think it's a terrific opportunity for to us to secure this arrangement with Delta for another two years, and then have the extension beyond that, that really exists at our option, going into a further relationship with them on what we believe are terms that are just terrific in today's environment. As you know, this whole business has gotten very competitive over the last few years, and to be able to extend this arrangement on these kinds of terms we find to be very attractive.

And the total economics of our business relationship with Delta, if you look at our Delta SkyMiles customers is absolutely outstanding, it's one of our very best programs with terrific returns.

So I don't think I can promise that you're going to see a cost reduction, but I do think you're likely to see a continuation of very attractive financial returns from our relationship.

And how will the purchase of miles be reported? So you're going to pay them $500 million over two years?

No, I'll give you an example. So 500 million of this is a loan. They'll be given $500 million basically in cash, I'm simplifying it here a little bit because it actually comes in a couple of different tranches.

But they'll be given $500 million in cash, and then there'll be no prepayment to them for a period of 24 months. I'm sorry, for a period of 12 months. Then during that time period from month 13, then the following 24 months we will, in a sense, amortize that $500 million up-front payment.

So let's assume just for talking purposes, that we would typically pay them $500 a month, and that $500, I'm sorry, $50 million a month, and that $50 million would go for some Delta SkyMiles points that are earned when people use their SkyMiles card. It would also come when people took their Membership Rewards points and moved them over to Delta.

There's two different ways that that can be earned.

Let's say in a typical month that we might pay them $50 million. Under this agreement they'll be paying us back, say at $20 million a month over a 24-month time period, so instead of paying them 50, we'd only be paying them $30 million and we'd be amortizing the 20.

Okay. Thanks.

You bet.

Sir, your next question comes from the line of Bruce Harting with Lehman Brothers.

I was afraid I'd be next. I'm still trying to digest, I didn't quite understand what you just said with the 50 million.

Do you want me to repeat it, Bruce? I think it's important. Let me say it one more time, because it is important.

So we have an arrangement with Delta that we will pay them $500 million up-front for prepaid points. That will actually happen in two separate tranches. We're not paying it all at once but two separate tranches separated by some amount of time and contingent on some financial performance on their part before we make that payment, so we'll have essentially a prepaid asset that will be sitting with them of $500 million.

Beginning with month 13, and again, just to use the example, these are not, by the way, the actual numbers, but just to use an example. If we had typically paid them $50 million a month for Delta SkyMiles points that are earned when people use a Delta SkyMiles card, and for Membership Rewards points when they are moved over into the Delta SkyMiles program, instead of paying them the 50 million, we would now be paying them, say, $30 million, because 20 million would be amortized over those 24 months, basically. So we'd just be paying them less than we would have paid them before.

Okay. Got it. Thank you. My question is, you know, I know it's very, very early, but any comments on the MBNA relationship, you didn't really talk about in that terms of refreshing our guidance or anything you want to say now that it's actually up and running and this is really your first quarterly conference call. That's the real question.

But just on this Delta, if you look at from the a scenario analysis, you know, I was talking to our airline analyst today and he said this is really a huge event, I mean the stock was up 17%, and, you know this really could bring the pilots to the negotiating table on a much more serious way. You did say that this deal is dependent on that, correct?

And you said it was also dependent on two other things. And I'm just curious, if they avoid bankruptcy, that's fantastic, obviously, terrific. Just in terms of a worst case scenario, God forbid if the airline ever had to go to bankruptcy, I heard you say you would do, you know, a different form of debt.

Right.

Debtor in possession financing.

Correct.

What would just be a contingency plan, if anything, if they ever went into, God forbid, liquidation? That would be the question. Thanks.

Let me not talk about Delta specifically in those terms, but let me talk more broadly about the experience we've had because obviously we have been working with other airlines who've gone into liquidation in the past. It doesn't happen very often but it happened with Eastern and with TWA, it happened with Sabina more recently in Belgium.

Basically, we generally have agreements with airlines that allow us to hold back a certain amount of cash, and depending on the health of the airline the amount of hold-back that we have increases over time. So if an airline starts to move toward liquidation, generally we're in a position of not being out a lot of cash to that airline right before the airline actually liquidates.

In the case of Sabina, for example, which is our most recent case, when all was kind of said and done there, we actually owed the remainder of Sabina, after it was liquidated, $19 million because of kind of the balance of the hold-back in payments that we had made.

And so I think our risk management people are just very attuned, and the agreements that we have with airlines are very attuned to helping us manage through these very difficult periods. So we have always kind of been able to work through that and have never had an issue of any magnitude there.

If an airline goes into bankruptcy, as you know, we've got lots of experience with that, that typically has not affected our relationship with the airline whatsoever.

Our MR programs continue just fine. We have a relationship with U.S. Air, we have a relationship with Continental. Those programs are strong, many of our Membership Rewards points are redeemed on those airlines, and so we are really quite confident about this arrangement being a strong commercial success for American Express.

Our take on this is, is that we're helping a partner that we value highly and in return for that, we're getting an extension of our relationship that will take us into the next decade and that's a very good trade-off from our perspective given our understanding of the risks associated with this industry.

As far as MBNA goes, over the next while my guess is there's going to be news that we're going to be able to share with you, but let me not let the cat out of the bag on that. I think we feel, we continue to think of them as excellent partners and we're just very enthusiastic about our working relationship with them.

Thanks, Gary.

Your next question comes from the line of Joel Houck with Wachovia Securities.

Thanks. Good afternoon. Help me understand a little better the MBNA relationship. I mean analysts have kind of various different estimates of the impact next year and most are saying maybe 5 to 10 cents. It seems like you're taking on a lot of risk for a small near-term benefit, and I understand that perhaps the longer term benefits are greater. Can you kind of talk conceptually about this thing over the next three to five years and what's so compelling about this in terms of potential return to shareholders?

Sure. First of all, Joel, we have a fair amount of experience with this in other countries. In countries where we've got significant business positions, you know, places like Singapore, for example, or Australia, places where we really have done business for a long time, where we have an established market position and we showed in a presentation to the analyst community a couple of months ago that in markets where we even had a previous proprietary business, and we added to that a partner through our network business, that both the proprietary business grew and the network partner grew, that the combination of those two things ended up increasing the total size of the pie.

And our fundamental belief here is that an open network over time is a superior form of doing business. It creates more competition, it will make our proprietary business stronger and cause us to be even more innovative than we've been in the past and we think that's good.

So we both have been able to calibrate the risk from our experience outside the U.S. and believe the competition in and of itself is good. There's very little capital for us required to participate in the network business.

In terms of hard capital, you know, there's not much systems development cost required, we already have the terminals in place. It gives us reasonable, we have good coverage already today so no need to potentially supplement that.

And the risk is, obviously, that we will face some type of cannibalization. And my guess is there will be some cards lost somewhere, but on balance, our experience has been that the total size of the pie increases, and it increases at very attractive returns.

The other thing that is important for us is always to have relevance with the card. So it's important to have high-spending card members, but it's also important to have lots of those high-spending card members.

And the more people that come into a service establishment wanting to use their American Express card, the higher the probability is that we're going to continue to grow in our coverage which is a key competitive factor for us. So there are just lots of reasons from economic to coverage to, from a customer perspective that we think this is a very sound business decision for us and is likely to be good for our shareholders down the road.

Thanks, Gary.

Your next question comes from the line of Mike Hughes with Merrill Lynch.

Thank you. Good afternoon.

Hi, Mike.

You mentioned a new re-engineering effort that was going to offset the gain of the [BENDER] finance business. Can you detail that a little bit and what kind of opportunity it may be?

Yeah, we did say that. As you know, we have a very active re-engineering program and at any point in time, we really do have things teed up that we would execute on if we had an opportunity to do so.

And I think you know, as we try to look at our long-term financial goals and ensure that we work towards these long-term financial goals, and if we're in a position of exceeding them, then we like to make investments that ensure that we can continue to do that long-term.

And so rather than take the gain that we have just received, or will just receive on the sale of this business and have it drop through to the bottom line give the overall momentum in the business, we think it makes sense to take some of these activities that frankly we've had on the shelf for a while, we'd had an abeyance for awhile, and go into those into the fourth quarter because it gives us margin opportunities next year that we wouldn't have otherwise had that will be good for our ability to invest. Obviously we know what those programs are given that we're already into the fourth quarter, but I think it's a little premature for me to characterize those on the telephone.

Okay. Thank you.

Sir, your next question comes from the line of Eric Wasserstrom with UBS.

Thanks very much. Gary, the credit losses in the quarter were very, very good and yet provisions came down to some extent which suggest that your outlook for future credit losses are also good. Can you just give us a sense of what the parameters are that you look at that make you think that perhaps in fact there's further improvement to come?

Sure. One of the key things I think to think about, Eric, is the investment that we're making in Membership Rewards pays back directly in improvement in provision, and we say this a lot but it really is hard. If we could disclose the information even more clearly than we do, I think, you know, it would be clear that when we're making these investments in rewards we really do see a payback both in terms of average spend, the speed of pay that we get on the charge card, but very importantly, in terms of the credit performance that we see.

It's interesting outside the U.S. as we have really pushed Membership Rewards hard over the last couple of years and made the program more relevant, the credit losses have just improved in lock-step with that, and so there really is substantial improvement that we see as a result of the investment that we're making in the rewards program, and we expect to get that kind of payback because of the investment that we're making there. So that's getting better.

We also look at the overall state of the economic environment, we look at what new unemployment applications look like.

For example, we obviously have a, you know, quantitative algorithm that we use to estimate these. We do that, we make a point estimate, and then we have three or four adjustment factors that we look at but generally look at the economic environment and we make some decisions to move this up or down within a narrow corridor based on what our call of as how we think the economic environment is going to evolve. And we review that rigorously each quarter.

So basically, it's the combination of, you know, what I think is truly outstanding risk management capability that have been developed here over the last few years, ongoing investment in our Membership Rewards program, and our current view of how things are working economically that we feel very comfortable.

And as you rightly said, if you look at the coverage ratios, 160%, 130%, we are in a very comfortable position relative to historical levels in both charge and lending.

If I could just follow-up on the economic element of it, can you just in very broad strokes give us a sense of what your economic view is? You view the environment as being basically benign to credit [inaudible]? Moderately favorable? Moderately disfavorable?

Well you know, our best indicator is always what's happening in small business, and small business was up 18% in the quarter. We always say that our historical experience has been that small business is our best indicator of what's happening in the economy.

When things start to go bad that spending turns very quickly South, and when things start to get somewhat better, it starts to turn up. And it's been running against easier comparisons earlier in the year up 22, 23, 24%, now we're up against a tougher comparison in this quarter and it's up 18%.

So honestly, through the third quarter, we've seen no reason to feel like there's been a substantial weakening in the environment. We're up against a little tougher numbers but overall the environment seems to be pretty solid.

Thanks very much.

Your next question comes from the line of Laura Kaster with Sandler O'Neill.

Good afternoon. In looking at AEFA, pretax margins have remained around 15% for the last couple of quarters down from 18% in the first quarter. Do you think that margins are going to remain stable at these levels given the fact that management and distribution fees as a percentage of managed assets continue to decline quarter-over-quarter, or do you think that from these levels you could see possible margin expansion?

Well, one of the wild cards, frankly, has been, you know, legal costs. As we mentioned right at the start of the supplement there that we've had higher expenses related to regulatory and legal matters at AEFA. And while we haven't obviously split that out in detail, it's obviously been a significant issue for this industry during the course of this year and that's had an influence on what that margin levels is. So if you take some of the underlying factors that are impacting our margins that are not exogenous, I mean things that are really right in the business, that there aren't really many trends one way or the other.

So, for example, in our certificate business, we've had a little bit of a margin compression as we have come through this year, but that is generally offset, as interest rates go up, with increases in spreads in other parts of our business. So from kind of an underlying perspective, the margins there have been relatively stable.

We have had a re-engineering program going there to make sure that our costs have been contained from an operating expense perspective, but the real differential in the time period that you're talking about here is the expenses associated with both complying with questions that we've gotten from regulators and then obviously the legal fees that have come out of various regulatory matters that they've had.

Do you have any visibility on improvement in the management and distribution fees as a percentage of assets or is that going to be stable for the time being?

Well, there's a couple of factors that are impacting that. One is that as I mentioned in the earlier call, with Threadneedle being a larger and larger portion of our total, those institutional assets don't have quite the same margin associated with them as others.

Also, we're coming from a point when AEFA was a closed distribution system, 65, roughly 65% or so of everything that AEFA had in its long-term mutual fund base was proprietary. And since we opened the network a few years back, that has changed. Typically in the industry an average, you know, 25% proprietary sales mix would be not unusual.

And so I think, you know what you're seeing is the compression associated with our moving over time to an open network system, and so that pressure is unlikely to change over the near-term, and obviously it has to be offset by growth in the broad range of activities that we're engaged in as well as our continued re-engineering activities.

Okay. Great. Thank you.

Your next question comes from the line of Brad Ball with Prudential Equity Group.

Thanks. Hi, Gary.

Hi, Brad.

Just a quick clarification. In terms of the regulatory expenses at AEFA, could you characterize a portion of those as one-time or are those ongoing higher costs that you'll just have to incur in running the business?

Honestly, I wish could I answer that question, but as you know, the discussion that we have with regulators are ongoing and it's difficult for us at this time to predict the outcome.

Okay. My question really has to do with the management of interest rate risk in the current rate environment. How, first of all, how you go about hedging the volatility of rates and the change in the slope of the yield curve, and then how should we think about that hedging cost as we look out to '05 in light of the economic comments that you made earlier?

The way we do it is, you know, in periods when the yield curve is flat, and relatively low, we obviously try and fund as long as we can, when it is steep and high then we fund shorter. And that's a short explanation, but let me tell you how it actually works operationally.

So in the last couple of years we've obviously been in a very benign interest rate environment and while we were in that environment we tried to lengthen out our maturities as much as we could. While we were doing that, we were actually paying more for that money than we would have paid had we been funding short. Because if we'd funded short, it would have just been lower cost.

And so we have hedges, and I've described the proportion of our business that is hedged in '05 and '06 on previous conference calls.

Now we're in a rate environment where, at least if you believe the yield curve, rates are going to go up in the future. As rates go up, we're going to be funding shorter rather than at the longer end of the curve.

So paradoxically you now might be funding at more or less the same rate, but because you're funding at a different maturity, you're actually coming in at a blended cost that is not all that different.

Now, it doesn't work out as perfectly as I just described it. In periods of rising rates we will see rates go up on average, and in periods of declining rates, we'll see them go down on average. But basically, those are the parameters that we try to manage in and we push out the hedges during those time periods when we have the opportunity, and then we pull back in time periods where we don't.

For right now we feel pretty comfortable this year obviously is well hedged. We've said in the past that '05 had strong hedging positions and we also had hedging that goes out into '06 as well.

So we feel, you know, given the other things that go along with an improving economic environment that our hedge position is probably going to be just fine.

I think you said in the past that you sort of set those hedging parameters early in the year, but what you just described sounds like you're doing it on an ongoing basis.

I think what you probably heard me say is for foreign exchange hedging we do exactly what you said, so we lock in our foreign exchange, we basically are a dollar reporting company.

We lock in our foreign exchange hedge just towards the end of the year, we just ride with those throughout the year. But from an interest perspective we look at those every day of the week.

If you to have close out a hedge because of the different rate scenario than you had forecast, where does that hedging cost show up? Is it in the net interest margin and in the discount rate for the charge card business?

Net interest margin, right.

Okay.

Does that answer your question, sir?

Thank you.

We'll now go to the line of Michael Hodes with Goldman Sachs.

Good afternoon. At this point a lot of my questions have been addressed, but, Gary, I was hoping you could give us a little bit more color on marketing and promotions line within TRS in terms of how you're thinking about the outlays in that specific line item as we move through the fourth quarter into next year? I believe on a prior conference call you made a specific comment talking about continuing to keep that at fairly robust levels.

Yes, I would characterize this quarter as robust. I think off the top of my head it was up like 29% in this quarter and I think we were up on top of 23%, if I recall, from the prior year. So it obviously is up and continues to be pretty strong.

The amount that we spend there is heavily driven by what we're able to accomplish from a re-engineering standpoint, and the results that we're seeing in the business as a result of it. And right now we feel, if you kind of look at the metrics of the business, bill business is strong, cards in force are coming through well, our average spending is progressing nicely.

All of the things that we would want to see responding to and provision is coming down. All of the things we would want to see responding to an investment in rewards and an investment in marketing is happening.

I think as long as we have the financial flexibility that re-engineering can provide for us, and one of the reasons why we talked about additional re-engineering in the fourth quarter is because that's going to be necessary for us going forward, is that our expectation is we'll continue to spend at these levels, or I don't know if they'll be exactly at these levels, but continue to spend at relatively high levels going forward as long as we see the benefit in our metrics.

Okay. And just a clarification regarding Delta. The $500 million, should we think about that just in terms of what normally would be flowing from American Express to Delta in terms of competition for rewards-related programs? Should we think about it in terms of a two-year time period or is this a subset of what would typically occur over two years?

The term is not typical necessarily of any time period. The term of this is for three years. In the first 12 months there's no payment that is made by Delta to us for this.

In the time period from the end of the first yield til the end of the third year, then we will be deducting from what we otherwise would have paid Delta, roughly $20 million a month in order to pay off the $500 million in prepayment. So it really is just a normal part of our business activity.

The thing that's unusual about it obviously is that we're doing this prepayment. But, you know, we have a real good idea about at what rate our Delta program consumes these miles and we feel very comfortable with the parameters of this program.

There's two aspects of this that make it a little complicated but I think are worthwhile thinking about. I mentioned before that when a customer uses their Delta SkyMiles card, that mile we pay Delta for immediately when it's used. When a customer, we only pay a customer for Delta SkyMiles out of Membership Rewards when those miles are actually moved into the Delta program.

And so there's two sources from which Delta gets paid basically, and we have obviously a good sense of what that is. It doesn't bear any resemblance to the $50 million example that I just used, but that's a simple way to think about. And some subportion of that, we'll continue to pay them and then we'll deduct the rest as payment for this loan.

And just lastly, given the magnitude of the rewards-related outlays, why don't you break that out separately? It seems like there are other line items that you do break out that are of comparable magnitude.

I'll tell you the way we've honestly thought about is it that rewards and marketing for us are so inter-related, there are sometimes when we're driving advertising, there's sometimes when we're driving direct mail, there's sometimes when we're doing double rewards points, and we vary these things and we don't think about them as different kinds of programs, we think about them as our arsenal of market expenditures.

In the past when we have split this out, you probably recall that about a year ago we split it out and it was roughly 50/50 and we did that for a quarter or two. We found that it was just an arbitrary distinction that really didn't reflect the way we actually managed these factors internally. So that's the reason why we are where we are.

Thanks a lot.

Your next question comes from the line of Bob Napoli with Piper Jaffray.

Good evening. How are you?

Good.

Couple of questions. Is the airline charge down 7% worldwide?

Yeah.

Did something unusual happen there? I mean it was up, [inaudible] comparisons got a little tougher.

There really is just a reduction in the average charge for airlines, so what you're seeing is good transaction growth, transactions kind of returned to the levels that we saw back in 2000, so there's a lot of us flying out there, but we're flying at a lot lower price than we used to a couple years ago, and that's what you're seeing in those numbers.

Okay. If you look at your portfolio, I mean, you talk about the economy and you feel pretty good about the economy looking at your small business, but if you segment your portfolio in the consumer side are you seeing any areas that have more weakness relative to others or more strength relative to other areas? I mean, because what you're saying is kind of directly opposed to what seems to be the general market perception that the economy seems to be softening. But you're saying you see no signs of that? I'm just wondering if you see within your portfolio some areas that maybe have softened somewhat?

Well, you know, if you, again, it's a little bit hard for us to judge because if you look from year-over-year, in the first half of the year we had the SARS/Iraqi war period and we were up against a very weak base, and now we're up against a much more difficult phase.

And to say we have the sophistication to be able parse that out that exactly would really overstate what our ability is to do. What we know for sure is that the economy is not falling out of bed at least as far as American Express is concerned.

You know, we track the sales by channel so if you take warehouse clubs and supermarkets and normal shops, we track each one of those categories of trade pretty closely and it would be hard for us to say that we detect any significant weakness in any of those categories.

There clearly have been times, if you go through the last quarter, where there was some change in pattern. So at retail, for example, during the month of August things appeared to be a little weaker. That appeared to bounce back a little bit during the month of September, and so we do see those kind of quarter-to-quarter patterns.

But within the noise level, you know, a percent our two of what we're able to actually calibrate these numbers it would just be splitting hairs to say that we've been able to identify something that gives us either large concern or any different feeling than we had a little earlier in the year.

Okay. And just a follow-up on the marketing expenses and if I'll get any more color out of you or not on this, but, is there, I mean certainly there's a significant percentage of that is directly tied to how much billable business there is per quarter. And what I'm trying to understand is how much of the marketing that I see is opportunistic because you have very strong earnings power and you're looking at opportunities to develop the future business versus how much of it is really directly related to the level of business we're seeing today?

Well, we do do exactly what you say. What we do basically, without putting too fine of a point on it is, we put together a marketing plan obviously for each quarter. That account for the majority of our marketing expenditures. Pick a number. Might be 70 to 80% of what we do. It's pretty much a programmed plan.

We then, as we go into the quarter, we look in, is the economy doing better than we thought, is our business performing better than we thought, do we have the flexibility to potentially spend more? And we've worked hard over the last couple of years to be able to flexibility increase our marketing on a rapid basis so that we could do new programs.

And by the way, this same capability works exactly the same way in reverse. So if we came into the quarter and we saw that things were weaker than what we anticipated, we've worked hard to have the flexibility to cut out spending also relatively rapidly.

And sometimes we have different directions in different businesses in the same quarter. We've had times when corporate has been doing poorly and consumer's been doing well, and we've made different choices in the same quarter.

So most of what we commit is pretty well committed at the start of the quarter but we do have a certain amount of flexibility that we retain that in relatively short order, I'm talking about in a matter of weeks, we can get into the marketplace and use it opportunistically.

And our use of that is driven by how we're doing financially. How we're performing relative to our target, have we, is our revenue growth rate at the level that we're comfortable with? Are we kind of meeting or exceeding those targets? And we do make those decisions basically as we go. I think that's what you're seeing reflected in these numbers.

And again, now because we see a bit of an opportunity to do some re-engineering that we might not had seen had we not had this gain that's coming in the fourth quarter, we may take advantage of that and hopefully that will provide some additional marketing capability for us as we go into next year but we'll kind of play that out. I think we're getting down probably to the last question here.

Thank you.

Your next question and final question of the evening, sir, comes from the line of Matthew Park with A. G. Edwards.

Good afternoon. Just sort of a cleanup comment on the credit quality outlook and what you have seen throughout the quarter perhaps on the business trends in general and credit quality specifically. I was wondering whether you have a lot of confidence that net credit losses can come down a little further on the lending side and whether you remain comfortable with the reserve on the charge card side?

And then on the business fundamental side, have you noticed any sustained rebounds or somewhat of a rebound from the summer weakness that you sort of alluded to a little bit in the previous comments, whether that's widespread or whether that's concentrated in a small business and the consumer continues to be a little weak? Thank you.

I think we really have had just excellent credit performance in the third quarter. Obviously I can't characterize what our expectations are going forward, but the reserve levels that we set, you know, that you can see in the numbers here, are a good characterization of our best guess at how we view the future.

We calibrate those things pretty carefully at the end of the quarter. So performance in both charge and in lending was really outstanding. It's industry leading.

As you know, if you kind of measure industry performance, it's industry leading and we obviously have a little tougher measure than most companies in the way we view these things. And even with that tougher measure our performance is leading the pack and continues to improve and to the extent we continue to make the investments in Membership Rewards and investing in our underwriting capability, we hope obviously, that that trend will continue for us.

We did see the rebound that I talked about a little bit from the August-September time period. Again, it wasn't a dramatic weakness and a dramatic shift of strength, but there was clearly some difference there. It might have been related to when the Labor Day holiday started and how that tied into the school year.

That kind thing, I mean it's that kind if issue probably that may have been in play there, but we don't see any fundamental change in our business as we've come through the last three quarters other than the fact that we're up against more difficult numbers now than we were earlier in the year.

So thanks everyone. We appreciate you being part of the call. As always, Ron and Company will be here this evening with any cleanup questions you might have and we appreciate you listening.

Ladies and gentlemen, we do appreciate your participation in this evening's conference call. This does conclude our American Express third quarter earnings conference call and you may now disconnect.