美國運通 (AXP) 2002 Q3 法說會逐字稿

Good evening ladies and gentlemen.

Welcome to the American express third quarter earning release conference time.

At this time all participants are on a listen-only mode.

Later we will conduct a question and answer session.

I would now like to turn the call over to Mr. Ron Stovall, Senior Vice President of Investor Relations.

Mr. Stovall, you may begin.

Okay.

Thank you very much.

And welcome to everyone.

Appreciate all of you joining us for today's discussion.

My job, as usual, is to remind you today that the discussion 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, planning, will, should, could, likely" and similar expressions are intended to identify the forward-looking statements.

And factors that could cause actual results to differ materially from the forward-looking statements including the Company's financial and other goals are set forth within today's release, which is has been filed in an 8K report with the SEC and in the Company's 2001 10K report which was already on file with the SEC.

As with recent conference calls this call is being broadcast live on our Web site.

Gary Crittenden, the Executive Vice President and Chief Financial Officer of American Express, will provide introductory remarks highlighting the key points related to today's announcement.

Once he completes his remarks, we'll turn to the moderator who will announce your opportunity to get into the queue for the Q&A period.

Up until then, no one is actually registered to ask questions.

So at that point and time you need to actually register and push the button.

While we will attempt to respond to as many of your questions as possible before would end the call, we do have a limited amount of time, so we would just ask that you limit yourself to one question at a time during the Q&A to give all the folks in the queue an opportunity to participate.

With that, let me turn the discussion over to Gary.

Welcome and thanks for joining with us today.

As you have already seen, our third quarter diluted EPS was 52 cents versus 22 cents last year.

Overall, the third quarter results demonstrated the continued flexibility within the Company's expense base.

The results from our focus on reengineering and improvements within the Company's risk profile.

These benefits coupled with lower funding costs allowed us to deliver solid financial results while absorbing three things: first the impact of a weak economic and market environment, second, significantly higher costs related to marketing and other business building initiatives, and third, increased expenses related to revisions of some of the factors underlining AFA's deferred acquisition cost.

Companywide our restructuring activities are on plan, and we're making very good progress here.

The consolidated headcount reductions of 13,800, or 16%, since year-end 2000 provided substantial cost benefits throughout our businesses.

In addition, our consolidated balance sheet is strong.

As usual, I will start with the headlines in each business then go into detail on each of the segments.

At TRS, the headlines are business volume improved versus last year on relatively stronger consumer spending and increases within the small business and corporate customer segments.

These better volume comparisons do not reflect sequential improvements in the underlying spending levels, but more so they reflect easier comparisons to the particularly depressed levels after 9/11 of last year.

On an absolute basis, we still have some distance to go to get back toward our historical growth level.

Lending spreads at funding cost continue to be a strong positive factor.

However they were less been beneficial than earlier in the year.

As expected, slower cards in force and lending balance growth reflect the impacted of our reduced new customer acquisition spending in the second half of last year and in the early months of this year.

The more recent increase in new customer acquisition spending you have seen is consistent with our guidance last quarter.

We expect to further increase spending in the first quarter.

As a result, we believe we will have positioned these metrics and TRS's's revenues for stronger growth coming out of 2002 and into 2003.

And overall credit quality continues to be well controlled.

At AFA the headlines are: The execution of our business plan is on track as reengineering benefits continue to fund the cost of upgrading or asset gathering and investment management capabilities, while equity market levels negatively impact results.

We're encouraged by the early progress being made on improvement investment progress, as again this quarter equity fund performance on an asset-weighted basis was above the industry mean.

In addition, overall sales levels and advisor productivity were relatively strong in the quarter.

But the persistent equity market volatility does pose a challenge to asset and sales trends over the near term.

In fact, changes to some factors related to AFA's [DAK] including a reduction of the assumed market returns going forward resulted in an incremental $44 million of expense during the quarter.

And finally American Express Bank continues to perform well in light of the difficult global economic environment.

With that, let me now review the details of each segment and I'll sart with TRS.

Revenue increased 5% and net income more than doubled on a reported basis.

Adjusted to exclude last year's restructuring charge and the disaster recovery cost and waived customer fees, net income rose a healthy 28%.

On the revenue side, billed business was relatively strong up 7% as U.S. consumer spending grew at 10 percent.

Small business spending rose 5%, and corporate volumes increased 2%.

The discount rate declined two basis points from the second quarter of 2002 and four basis points from the third quarter of last year.

This was driven by the mix of spending from relatively stronger growth in everyday spending categories.

Worldwide cards in force rose 3% versus last year, although the impact of lower acquisition spending has tempered the sequential growth here in the U.S. during 2002, outside the U.S.

Growth remained relatively strong at 7%.

Lending-spread revenue grew 10% on 6% revenue growth in quarter-end lending balances and substantially higher yield versus last year.

Travel and travelers checks sales remain depressed and continued weakness in travel activity.

Marketing expenses rose 31%, reflecting the introduction of our new charge cards with membership rewards built in and our new cash card.

Additionally, we spent more on loyalty marketing and renewed our focus on card acquisition activities.

It's worthwhile noting here as well that we increased our marketing expense despite the reduced level of securitization regained versus last year.

Provisions for losses declined 6% as overall credit quality improved.

Reserve coverage of lending receivables was increased in light of the continued uncertainty within the economic and regulatory environment.

With regards to the banking regulators, as I indicated last quarter, we are in regular dialogue with them on an ongoing basis and have not been alerted to any particular issues with regards to our portfolio or practices.

However we do think in the current environment it is prudent to maintain a relatively conservative reserve posture.

Coverage of charge card receivables was reduced somewhat based on the portfolio's excellent credit trends.

Despite this, coverage of past due balances remains at a very high level.

As forecasted, interest rates vest was sharply lower on a significantly lower average funding rate, and to a much lesser extent,t lower receivable balances.

Human resources expense declined for the fifth quarter in a row on continued headcount reductions.

The TRS employee base was down 11,300, or 15%, versus last year's level.

Other operating expenses rose 15% versus last year from higher participation in our rewards program and the impact of our technology out soaring agreement with IBM.

As you know the IBM agreement has the affect of moving certain technology related cost from human resources expense to operating expense, although at a lower overall cost.

Now let me turn to the details at AFA.

Reported net income rose 5% versus last year.

However adjusted for last year's restructuring charge and disaster recovery costs, earnings declined by 21%.

This decline includes the $44 million [DAK]-related expense increase that I referenced earlier.

Revenues were flat, as you were expect in this market environment.

Management and distribution fees declined 7% on a lower level of average assets under management.

The lower asset base reflected the negative impact of market depreciation, and to a lesser extent the negative [AFA] outflows within both our retails an institutional management activities.

The lower fees earned on these reduced asset levels more than offset the fact that distribution fees rose versus last year on solid sales growth.

For example, we saw improved sales comparisons within both the retail and institutional channels.

Advisor productivity was relatively strong, as Branded Advisor-generated sales increased 9% on a cash basis and 8% as measured on the internally used growth dealer concession basis, which waves the sales of various products to reflect their individual profitability dynamics.

Other revenues rose on greater life and property casualty insurance revenues and higher advise service fees.

Human resource expenses declined 3%, as higher field force compensation related cost due to the increase sales levels were more than offset by the 14% reduction in the average home office employee base.

The advisor base was down slightly versus the second quarter of '02 in last year.

We continue to see the effect of the difficult market environment, as well as our more conservative stance towards advisor additions and attrition in light of the weak environment.

The veteran advisor retention rate however remain strong.

Other operating expenses rose 40% versus last year, due primarily to [DAK]-related increases in expenses.

In addition, the increase reflect the impact of the IBM outsourcing agreement and a higher minority interest expense related to premium deposits in AFA's joint venture with American Express Bank.

With regards to DAK, during the quarter AFA completed a comprehensive review of its DAK-related policies and procedures.

This review which included benchmarket assistance from an industry specialist, analyzed various historical AFA/DAK dynamics in addition to industry benchmarks.

The specific areas reviewed include cost deferred, customer asset growth rate, reversion to mean assumptions, DAK amortization periods, mortality rates and product persistency experience.

As a result of this review AFA recorded a net incremental $44 million in expense during the quarter.

This expense reflected revisions related to three key drivers of DAK expenses: first, customer asset value growth rate assumptions for variable annuity and life products, including revision-to-mean assumptions, were adjusted downward to 7% for the near-term and long-term.

While this growth level is fairly consistent with the prior long-term assumption it is approximately half of the prior near-term assumption.

This change resulted in a substantial acceleration of DAK amortization.

Second AFA revised assumptions for DAK amortization periods, predominantly for fixed annuities to better reflect actual experience and future expectations that.

This resulted in a substantial DAK amortization decrease.

Lastly we revised the types and amounts of cost deferred in types to reflect the impact of advisor platform changes and the related reengineering that we've done over the last year or so.

This represented approximately half of the combined expense increase during the quarter.

The revisions related to market and amortization period assumptions should reduce adverse DAK expense volatility going forward.

The changes regarding deferral of certain cost will also increase ongoing expenses somewhat.

This increase should be offset to some extent as reengineering and and other cost control initiatives are expected to mitigate its impact.

One last point regarding an important topic in the insurance industry.

That is, guaranteed minimum annual death benefits.

Unlike others in the life insurance business, the design of our Minimum Annuity Death Benefit guarantees generally reset up and down in six year intervals, not annually like other life insurance companies.

This leaves AFA less exposed to expense-related risk due to near-term market volatility.

As a result, we saw only a modest expense impact in the quarter.

Now for the details at American Express Bank.

The bank continued to make progress on its strategy shift.

Consumer activities expand as private banking client holdings and PFS client volumes both rose 9% versus last year, and corporate banking loans continue to decrease and now stand below $500 million.

Overall results continue to benefit from lower funding cost and good expense controls.

These benefits were partially offset by higher provisions on continued consumer losses in Hong Kong, where the environment has been particularly difficult but seems now to have stabilized.

In summary, that's a quick review of our results for the quarter.

Let me conclude with a few comments on our outlook.

Clearly, we are still dealing with the difficult economy and volatile equity markets.

Questions still exist about the likely direction of each over the near-term and the added uncertainty regarding the prospect of war in Iraq and other geopolitical events further cloud the view.

The flexibility we have now built into our expense base coupled with the benefits of good credit controls and lower funding costs allowed us to produce solid earnings during the first nine months of the year, despite the weak environment and the cost of additional spending on revenue-generating initiatives.

The results show we are now better positioned to operate in an uncertain environment, and as a result, we do expect to deliver solid results again in the fourth quarter.

At the same time we plan to further increase our growth-oriented investments, particularly in marketing.

These expanded business building efforts should lead to improved business metrics late this year and into 2003.

In view of that accelerated spending we continue to expect that earnings per share will not exceed the $2.01 referenced last quarter given current conditions.

But as always, we will carefully monitor the performance of the economy to endeavor to make the appropriate tradeoffs.

Thanks for listening.

We're now ready to take your questions.

Thank you.

You will now begin the question-and-answer session.

You have a question you will now need to press the 1 on your touchtone phone.

Remember if anyone pressed the one at the beginning of the call, the request has not been recognized by the system.

You will hear an acknowledgement that you have been placed in queue.

If your question has been answered and you you wish to be removed from the queue, please press the pound sign.

Your questions will be queued in the orders received.

If you are using a speaker phone, please pick up the handset before pressing the numbers.

Once again if there are any questions please press the 1 on your touchtone phone.

Stephen Eisman from Shipman Investment is on line with your question.

I can't believe I'm first.

Never happened before.

There you go, Steve.

Simple question on the DAK.

I'm somewhat of a newcomer in this in a little bit of life insurance.

Some of the companies that have recorded DAK impairments over the last come days have averaged about, I've been told something like 10% of the variable annuity DAK that they have on their books.

I am wondering if you can tell us what percentage of your variable annuity DAK the charge was, and it sounds like it was significantly smaller.

If it was, why do you think that was, compared to other people?

Steve, I don't have a good -- the total amount of DAK that we have on the balance sheet is $3.9 billion at the end of June .

How much would that related to variable annuities?

I don't have the split right in front of me.

But at $44 million it was significantly smaller than 10%.

I think, you know, each individual companys' circumstance is different.

One thing that is unique with AFA] as an institution is that we have had unusually strong product persistency as you probably know in virtual all 6 our annuity and insurance products; longer than what one typically finds in the insurance industry.

And what e we have gone through the progress of evaluating the DAK, we've looked at all of those parameters and are confident that we have drew them up appropriately.

So I think you have to look at the individual circumstance of each company to really answer the question in detail.

There were obviously some factors that were positive here for us.

Those tend to be things that were traditionally strength at AFA, and then we had the negative associated with the performance of the equity markets.

I guess one way to think about the fact we might be a little bit different is we do have a very sizable fixed annuity book as well which has DAK relating to it.

So one thing you could look at just to think about some of the size factors would be the fact that as of June, the separate account balance of $25 billion relating to variable annuity and the fixed annuity reserves were about $21 billion, so they're both of substantial size.

Okay.

Thank you.

Michael [Frudenstein] from JP Morgan is on on line with a question.

Please state your question.

Hi guys.

There's definitely a lot to like in the quarter.

Billed businesses showed resiliency, credit improved nicely, expense management was evident, you invested in market despite that.

Looks like you bought back about shares at about 60% of net income level you suggested in August.

Really the only issue I am still struggling with is the size of the DAk writedown.

And I guess by your own comments and the first question, it's clear we need to get a better understanding.

So just to make sure I understand your growth rate change; if the market is up 7% next year, I would assume you would not foresee the need for a DAK writedown to true-up assumptions.

That's my first question related to DAK, and these were all DAK related, Ron; so this is one question, okay?

The second is what is the cap on the reversion to the mean assumption you are using?

And then you also stated in the supplement and in your comments that you experienced a substantial decrease in DAK amortization related to fix income product.

So I am just wondering if that's prudent given that when equity markets improve, in theory, your persistency on fixed income products may weaken.

So if you could comment on those thoughts.

Thank you.

On that single question, Michael?

Yes.

The answer to your first question is yes, 7% if the market appreciates 7% next year.

And assuming all other variables were the same, there would be no DAK adjustment at all.

I don't understand what you mean about cap on the reversion-to-mean.

Could you give me more detail what you are talking about?

Typically, my understanding is that you have a cap something like 15% over a period of time in which if the average of the returns in the were not getting you to that overall 15% level that you might require another true up.

That's what I'm trying to understand what exactly your cap is.

Michael, we don't have a cap like that on our assumption base.

I know there are some people that have like a 15% assumption if you had a dramatic runup like that; we don't have a cap like that in our underlying mythology.

Then finally from a fixed income perspective -- or from a fixed annuity perspective, what we have done is we have taken our actual fixed annuity book which, as you might guess, you know has tenures years to go back many years and we have actually trued up the length of time over which we were referring DAK to reflect the actual experiences of the fixed annuity book.

This isn't a circumstance where we have just sold a lot of new fixed annuities, and as a result of that we're some way getting a benefit associated with it.

What we've really haven't taken the whole historical book and gotten comfortable that we are reflecting our actual experience in the amortization time period.

Okay.

I appreciate that.

If you could just tell us if there was any billed business pattern unusualities in the quarter.

Was billed business consistent throughout the quarter?

That's my last question.

There was because of the effect of the September, 9/11 event.

What we saw in the first couple months of the quarter is a trend that would have been quite consistent with the numbers that we saw in a full quarter basis last quarter, and obviously then given the impact that 9/11 had on the corporate business and small business business, we saw a big bump in those percentage growth rates during the last month of the quarter.

I meant month to month.

Oh sequentially?

No there's not a substantial sequential difference.

I'm sorry.

Thank you very much.

From Ken [INAUDIBLE] from Morgan Stanley is online with a question.

Please state your question.

Thank you.

I've got a question about hedging of the S&P 500.

There was all the commentary in the earnings supplement in the AP units and I recall in the past you all have hedged pretty carefully to management fees at some points in times and at other points in time, you all have elected not to hedge.

I guess my question is, can you talk about how much of a difference the hedging made to the results this quarter?

And then what is the posture going forward?

Is this a period when you are -- will get further benefits if the markets fall or have your hedges run their course.

We didn't have any hedges on the equity markets in the quarter.

So there was no benefit associated with it now.

And currently we're not hedging and don't have any near-term plans to hedge the S&P or other equity indexes for that matter.

It's something we review every month, and we look at the level of the equity markets and we try to make a decision whether or not we think it's an appropriate thing to do.

On balance given the equity markets at least historically have appreciated, it's generally been bad to bet against that long-term appreciation range.

So generally we haven't.

Okay.

The reference in the notes which you would find in really pretty much every quarter over recent periods talks about hedging relating to specific certificate products or there are some annuity products that are tied to the, S&P 500 rate.

Gary's point, the hedges don't relate to management fees on the underlying assets under management.

They really relate to the ups and downs in the S&P 500, and that affects what we pay the customers at any point and time; and we kind of perfectly hedged that as customers buy the product to offset it.

So any benefit or detriment that flows to investment income really gets reversed fully within the provision line, which reflects our payment to the client that either a higher rate if there's appreciation or lower rate if there's depreciation.

So there's really no pretax benefit; that's a hedge that's easily done, because you know exactly what was invested and at what point and time and what your cap is based on the contract with the client.

Okay.

Thank you.

From Fred Ball from Prudential Securities is online with a question.

Please state your question.

Thanks.

Gary, could you talk about your exposure your various exposures, both in TRS and in the insurance investment portfolio to the airline industry bankruptcies and how you've protected against the risk that may arise from the challenges of that industry?

Sure.

Obviously, we do a lot of business with the airline industry.

I think it was about 13% of the total billings in the last quarter, at least domestically.

So this is a risk we manage very carefully.

There really two or three different things that we do.

The first thing we do is we try to match as well as we possibly can the time period that we paid to the airlines with the time period that it takes for a typical customer to fly.

So let's say an average customer flies -- buys a ticket and flies within 21 days, particularly in cases of weaker airlines where we think there might be a substantial risk of bankruptcy, we would try and match as much as we can the time period in which we would pay them so that we don't have too much exposure.

Secondly, we have an airline that we have concern about, and we obviously work with that airline on a daily basis.

We might increase the amount of hold back that we keep on that particular airline.

To give you a good example, when [Sevena] went out of business earlier this year at the time that they went out of business we actually had a slight balance that we owed to them rather than them owing us to.

That being said we obviously still have sizable receivables with many airlines, some of which have had difficult financial circumstances; but historically as long as an airline has continued to fly, even after bankruptcy, we have been able to move from period to period and roll over that indebtedness.

It's one of the things the bankruptcy court allows to be rolled over, even in the event of a bankruptcy because being able to take credit cards is a necessary part of how the airlines do business and they couldn't operate without it.

So our position has been protected as long as an airline continues to fly.

The final line of defense, I would say, at least historically as airlines have gone bankrupt other airlines have picked up and honored their tickets because the airline industry of course wants to confirm the fact they will continue to honor tickets for customers that buy tickets in advance.

So all of those things taken together represent our risk mitigation.

Then we also maintain a reserve in our provision that reflects what we think is our best guess for losses in travel related industries.

Did you add to that reserve this quarter?

I don't think so.

We are in that reserve right now kind of one standard deviation above our projected losses.

And so there's really no particular reason at this point to add to it.

And then in the investment portfolio at AFA what's your exposure in terms of airline industry bonds?

We do have exposure there in particular with a couple of airlines that either recently have had or may have difficulty over the next few months.

It is not something that I would characterize as material.

There's two aspects.

There's the size of th indebtedness then also the degree of security.

We've reviewed carefully the quality of the security we have.

We think in the vast majority of cases we are in a well- secured position.

In those areas where we think we have exposure, we have recognized that and are thinking about the collectability on those bonds at this point already.

Thanks.

With Phil Hobson Wachovia Securities online with a question.

Please state your question.

Thanks.

I was wondering you know you had actually solid credit results.

Is there a particular reason why you didn't focus or highlight and could you talk a little bit about what's driving that?

It looks like the U.S. loan slowing a bit ,yet showing pretty good improvement in loss rate sequentially from the first quarter of '02.

Thanks.

It has been good performance.

And I think [Ash] and his team has done an outstanding job over the last year or so.

One of the things that is not to be found anywhere in the documents is this the fact that our performance on fraud has been outstanding.

That's one of the things that goes into our provision.

Ash and his team have continued to focus on that area this year, and it really has acted in part to offset some of the losses that we otherwise would have had.

This tends to move I think as you know very directly with unemployment rates.

As we've seen in moderation in the unemployment rate this year, we've likewise seen in moderation in our credit losses.

We would expect frankly that if unemployment stayed at the same level it would remain about in this range; if unemployment should deteriorate; and our planning assumption for next year is would we expect to see some kind of modest deterioration in unemployment, that it's likely we would see a modest deterioration linked the to that in our credit statistics as well.

The other thing I'd point out is that. particularly the charge side of our business where our charge receivables, so to speak have been slow growing for the first half of this year, the tenures we have in the charge portfolio are long and obviously very attractive, and that's one of the reasons why you are seeing the improved charge card performance.

I should add with the same breath as we introduce this new charge product that has the membership awards imbedded into it that we anticipate seeing growth in that charge portfolio.

As we do that, we're likely to see tenure-related deterioration in that ratio down the road, as well.

Was there any improvement because of lower incidents of bankruptcy filings in the quarter?

You know, the growth in bankruptcies has been pretty modest.

So it hasn't been a major quarter-over-quarter, year-over-year factor for us.

Okay thank you very much.

You bet.

David [Hodgkins] from Bear Stearns is online with a question.

Please state your question.

Thanks , I wonder if you could talk about the growth of cards in force, and specifically, what changes there might have been in account administration sequentially mentioned in the supplement I think about cancellation of corporate accounts from I guess employee layoffs.

But I wonder if you could talk about how the marketing spending might have contributed to the growth outside the U.S.?

Right.

The marketing spending has been more consistent through this year outside the United States than it has been in.

We are obviously more focused domestically on in our marketing activities now in this quarter and next quarter.

And we said last quarter we expected to be at our low point in cards in force growth rate in the third quarter that we should see some uptick in the statistic by the end of the first quarter and obviously we're hoping to see that continue into next year.

And so I do think what you are seeing now is a reflection of the spending levels we had in the this early part of this year as we very well very concerned about the overall level of the economy and what impact that would have on our financial result.

That being said you have also put your finger on a couple other important influences.

The attrition for example has impacted us in the corporate area as some major customers of ours go through layoff process and people depart then we obviously lose those cards and those are high spending cards for us.

Additionally, we have had some attrition that we have caused to happen in both Brazil and Argentina which has had an impact on our attrition and statistics internationally.

So if you just look at our average attrition because thof factors we've just talked through those numbers are up a bit from where we might typically see them.

So you are seeing both the impact of higher attrition as well as the fact that our marketing spending was particularly weak in the united states at the end of last year and the first part of this year.

And hopefully the things that you are seeing right now in on television and hopefully in your mailbox are the kinds of things that we believe will lead to improved metrics at the the end of the year and early part of next year.

Okay.

Then related to that, could you just comment on any foreign currency translation effect of building business outside the U.S. this quarter relative to the second quarter?

It is in the supplement.

It is about a 1% differential on a total company basis.

The outside the U.S. , we were up 8% on a reported basis for billed business.

FX adjusted would have been up about 5%.

So a little bit larger spread outside the U.S. than we had seen in a number of recent periods.

And actually it's kind of flipped over versus some of the quarters last year.

But generally, as Gary said, a pretty minimal effect on the overall built business worldwide.

Yeah, so total company 1%, outside 3%.

Thanks a lot.

We have Nat [INAUDIBLE] with Solomon Smith Barney online with a question.

Please state your question.

Okay.

Thanks.

Was just wondering, on the discount rate there's a line in the supplement that talks about selective repricing initiatives.

I was just curious if you could comment what that's alluding to?

We on you know virtually every month evaluate our pricing levels that we have in each of the different industries.

As you know, we price differentially by industry.

In some cases where we think our value proposition is strong, we might move that up.

In some cases where we think it's appropriate, we move it down.

These are never very significant changes in discount rate, but we do make selective changes like that.

That actually is the not the biggest influence though in the discount rate change; by far the biggest ill pact is the growth everyday spend versus the growth in the more traditional T&E-type categories.

Everyday spend I think reached 62% in this quarter as a percentage of our total, and that's the primary driver behind this.

But we also obviously go through all the time and review our pricing and make sure it's appropriate.

Has the increase in everyday spend verticals caused any sort of pricing pressure in any of the other verticals?

You know the honest answer is there is always pressure in all of these verticals; that exists all of the time.

And our job every day is to continue to strengthen our brand and to make our average spend levels attractive that we justified the premium discount rate that we have.

So we, you know, we benefit from the fact that we have a very strong branded product that succeeds very well in the markets in which we compete.

But that creates tension obviously.

We have those discussions consistently.

But there's nothing new about that.

Certainly nothing recent about that.

Okay.

Great.

Thanks a lot.

You bet.

Matthew [Park] from Thomas Weisel Partners online with a question.

Please state your question.

First, does your level of the provision on the charge card side signal some kind of an expectant or anticipated improvement in the [INAUDIBLE] rate; or how would you think about that in terms of a quarter sequentially going down?

The way to think about it is that obviously the amount that we currently have in the provision reflects the losses that we expect to incur related to receivables we currently have there.

That's the way that is done every quarter.

The number coming down is a very slight change relative to last quarter in terms of percentage coverage.

If you look historically, our typical coverage has been kind of averaging in the 130s or so.

Last quarter, because we were somewhat concerned with the general economic environment, we bumped that up into the 160s.

But frankly, our performance was so strong this quarter that we brought it down in the same percentage terms as the improvement in delinquencies had in the quarter.

So, you know, going forward our expectation is this will continue to be tightly linked to what unemployment does.

But we believe that we are very strongly reserved and that our balance sheet is strong.

Matt, I think to Gary's point, if you looked at the coverage of the past two accounts there at 161% in the quarter while it's down a touch from 164% in the second quarter, if you looked over the last five years the high point was about 142%.

So this is still substantially above where we've been anywhere in recent history.

Okay.

And in terms of comfort about that, does the fact that going forward perhaps more card holders would have the reward program built into our cards play some role in terms of improvement going forward?

There's two factors.

There's that one, which is a positive for us; that the negative one is the one I referred to a minute ago, that we're going to see earlier tenure growth in the charge product and it's likely that's going to result in some deterioration in that ratio going forward.

That's a positive thing because it means our franchise is growing and healthy there.

But clearly what we've seen I think you have seen the statistics before that we have much lower charge offs for those people that are part of the rewards program, and that's obviously the hope of this new product, that we'll see lower charge offs going forward.

Great.

I missed the earlier part of your comments so apologies if you clarified this already but could you elaborate on the Canadian lending card shipped - that's shipped a million and a half cards this quarter?

I'll go through that.

Basically the cards in force numbers that have been reported historically have appropriated include the one and a half million Canadian lending cards that we talked about adjusting within the basic cards in force calculation.

So really if you went back historically, it has become apparent to us that this 1.5 million Canadian lending cards should also have been included as basic cards in force historically.

So we have now brought them into the basic cards in force number outside the U.S. in the third quarter.

It really doesn't affect the cards in force trend.

It has a small effect on the overall spending per basic card in force numbers, which we've shown you the revised growth rates including those basic cards in both this third quarter as well as last year's third quarter.

So you get a sense for the overall impact.

But in essence those were cards that were reflected as supplemental cards, meaning cards in force but not in basic cards.

And as we see it now based on the fact that the stand alone relationships with customers, we are including them in the basic cards in force numbers.

Okay.

So I should --

Going forward, they'll be in there.

If you want to think about the prior year numbers, you could add in there that that 1.4 or 1.5 million cards have been present in the cards in force statistic but were not previously included in basic cards in force.

Gotcha.

Okay.

Great.

Thank you.

Mike Hughes from Merrill Lynch is online with a question.

Please state your question.

Hi it's been covered.

Thank you.

Thanks, Mike.

Robert Hutchison from Goldman Sachs online with a question.

Pleads state your question.

Just a quick question: You alluded to the guaranteed minimum benefits.

Could you highlight your accounting on it, and why are a six-year cycle versus a one-year cycle?

It has to do with the product we offer.

So that 85% of the insurance contracts that we offer have a feature that has the repricing take place every six years, and the pricing can go obviously up or down depending on what's happened during that cycle.

So in any one year, we'd only have one sixth of our portfolio repriced, so that is that gives you a much longer time period.

So if you have add run up as we did over the last few years, for the contracts that were existing during that time period you could have that run up and then with the drop in equity markets today you wouldn't have to be facing the higher level of equity markets that would have raised the minimum.

Is this just a more conservative way that you historically have structured the product?

Or is there a competitive benefit doing this it way?

Just a more conservative way that we've structured the product, so we account for it in the same way.

But it's just the product that's structured more conservatively.

I have a follow up on another question, but I'll rejoin the queue.

No, go ahead Bob.

I just wanted to talk about the fixed income markets.

Obviously, a number of companies in the financial services business have been affected by wider-spread market access, you know regulatory, developments.

And you know obviously American Express is a complicated company on a consolidated basis.

You know, you're obviously very well capitalized.

You have a relatively small balance sheet.

But I just wanted to get your general thoughts, Gary, you know philosophically how you would take advantage of what would seemingly be a competitive advantage in certain areas in financial services, and how you might be approaching you know these markets or any markets differently in light of the distress that's evident broadly in the fixed income markets.

That's a broad ranging question.

Let me just give you a few thoughts, and if I miss the specific point that you are headed towards maybe you could redirect me a little bit.

The first thing I would say is that you know we have done a lot to insure that our liquidy is sound over the last probably year or so now going back to September 11.

As you know we've cut our alliance on short-term paper.

We've pretty well matched off our short-term paper borrowings with the amount of backup facilities we have.

So I think as we stand today, we have a very strong liquid balance sheet with substantial backup capability, and even beyond that, you know clear action plans that we would take should there ever be a liquidity question that we would face.

We have also tried to do, in this environment, take advantage of the low interest rates that we see not only for this year but for next year and the year after that.

So in periods like this where the yield curve has been particularly benign and where the slope has been benign, we have taken advantage of that to try to lengthen out our maturity so that obviously we can benefit from the lower interest rates for a longer period of time.

So more than I think might typically be the case, we've got our position you know filled for next year and some of our position filled for 2004 and 2005.

That obviously gives us a base line of things that we don't have to worry about in terms of interest rate increases that might be a little bit different than it has been historically.

This environment has created an unusual number of opportunities for us to look at, you know business, lines of other companies, different types of acquisition opportunities; in general, our view remains the same: that we have a organic strategy that is succeeding very well, that we participate in two industries that we're very comfortable in, and we have plenty of growth opportunities internationally and domestically.

And while we may do selected, relatively small acquisitions compared to our total size, we neither feel the need nor see any compelling opportunities right now that would change our view on that.

That kind of covers the waterfront a little bit.

I don't know if I covered the topics were you specifically interested in.

That generally does it.

I mean, where would you be specifically extending you know with respect to the geography on the balance sheet, is it in the consumer lending business or --

Yeah.

Specifically.

Now on the owned assets side, I think we've got a clear game plan there and we've laid that out with respect to what our high-year-old strategy is.

But the thing I was referring is to was the consumer lending side.

Thanks a lot.

Eric [INAUDIBLE] from UBS Warburg online with a question.

Pleads state your question.

Thanks very much.

Gary, the investment portfolio kicked up significantly this quarter versus where it had been trending over the past several quarters.

Could you indicate perhaps what the driver of that was?

And additionally, the -- I know you commented here in the release on what the year-over-year trend was in yield.

But it seems that the sequential quarter -- the sequential yield actually improved in the quarter and if you could give me some color on why that was?

The driver on the investment portfolio, we really did have a very good quarter in certificate sales and annuity sales.

The result of that is we had cash to invest, and that bump is what you are seeing in the owned asset portfolio, the investment portfolio that we have.

I am actually on the yield, our yield sequentially on a percentage basis did drop down, I believe, ten basis points from last quarter.

It's up substantially from last year at this time.

I don't know the number off the top of my head but maybe 80 to 90 basis points relative to last year, but it did tick down just ten bays points from last quarter.

Thank you.

Jennifer [Scudy] from CIBC is online with a question.

Please state your question.

Hi.

Most of my questions have been answered but just a little technical question.

Could you talk a little bit more about the travelers checks sales?

They were up quite a bit in the third quarter rel relative to the second quarter.

I do notice that kind of looking at it from last year, we did see a similar type of rise.

It just seems that it's to a greater magnitude this year than in the prior year.

Just kind a little bit of color as to what's going on there.

Thank you.

There is a cyclical pattern that takes place between the second and third quarter that is just related to people traveling and kind of seasonal-vacation type patterns.

There's a couple underlying factors that have also played a role in the business this year.

One is that with the conversion to the euro, there has been quite a bit of demand for this product.

That demand was obviously more heavily skewed towards the early part of this year, where people were buying this as kind of historic value where they could convert out of what they might have had and into the euro and then they're holding these.

So our average days that we have outstanding in the portfolio have increased a little bit, which is positive for us.

That being said, this is a business which obviously is very mature business and has you know competitive issues relative to other product alternatives.

And our group here internally is working hard on trying to reposition that business for us so that we can hopefully get a brighter growth future than we've had over the last few quarters.

But generally the pattern you are seeing is one more of seasonality than anything else.

Okay.

If I could ask just one quick follow-up.

You touched on this earlier about the credit quality trends within the lending portfolio.

The improvement that we saw sequentially was quite substantial.

Is this the beginning of a trend, and could we see losses perhaps stabilize in the mid-5% range for this portfolio going forward?

Again I think it's most likely to be related to unemployment.

I hate to keep coming back to that, but the relationships are quite striking.

And so as we've seen some improvement in those statistics, we've seen improvement here.

That's been buttressed by what I said about fraud.

There's also been substantial progress made on the fraud front this year, which has brought that number down probably more than we would have seen otherwise, certainly more than we would have seen otherwise.

I think there are two factors at least that influence this going forward.

One is the level of unemployment.

The second is the new tenures that hopefully are going to be growing in the charge business with the launch of this new product over the last quarter.

I think that combination is likely based on our planning assumptions to result some increase in this over the next few quarters if we're successful with that product launch that.

That being said, I wouldn't read anything into that other than what I just said; that new products at early tenures tend to have higher loss rates than older more mature products, and this is a very tenured portfolio that we're look at right now in the charge business.

Great thanks for clarifying that once again.

[INAUDIBLE] from Credit Suisse in online with a question.

Please state your question.

Thank you.

I am just following up on a earlier question on the charge card reserve.

You mentioned that the performance currently was very strong and took the reserve down in dollars at least substantially and kind of roughly maintains it in ratio terms, but you also said you expected three things to happen.

One was portfolio to grow again.

Second was the portfolio to show some deterioration and at least as a result of that growth, and third, that you expect unemployment to rise.

So I guess in that context I found that to be unusual.

It would mean that you probably have to put at least that amount, that $100 million, or so back in the reserve in some future point.

If we weren't at this level of coverage, if you look at the coverage levels, as Ron cited, historically we've run typically in the 130% coverage rage.

We're now at a level that is well above that in the 160% or so coverage category.

As you know, we reserve based on the actual size of the portfolio.

And we --

Isn't it based on a prospective expectation of loss, not an historical one?

Sure.

But you can't reserve for receivables you don't have yet.

And we're a very conservative overall level of reserve relative to our delinquencies.

We bumped that level up substantially last quarter.

If you go back two quarters I think we were at 135 or so, Ron?

About 135% coverage.

Last quarter we bumped that up substantially, and again after we had two sequential quarters of very substantial improvement in our performance, we brought it down modestly and still maintained coverage at levels that are, by all historic standards, very high.

Thank you.

From Deutsche Banc is online with a question.

Please state your question.

Hey, Gary.

Getting back to the DAK for a second, the $44 million was the net of three things, so it really isn't the -- we really can't tell what the writeoff is compared to other companies.

So I was wondering if you could break out those three things so we have an idea just the writeoff of the DAK on the variable annuities was as opposed to the netting of it?

And then what are variable annuity assets?

On the first question that you had, Mark, as you said there were three pieces associated with this.

The ongoing piece of, you know the reduced amount of DAK that we're going to be expense that we're going to be deferring, that's roughly half of the $44 million something like that.

Then we had you know a negative effect associated with the reduction in the assumed asset growth rate that was a multiple of the $44 million.

Somewhere between 3 and 4 times the $44 million.

Offsetting that was an improvement in the fixed annuity in particular, actual life period that we had.

And so that offset down to whatever the $44 million range was.

Hopefully with those pieces you can kind of triangulate on the details.

I'm not too good at geometry.

What was the writeoff of the DAK on the variable annuity assets?

Well we didn't split that out specifically.

Let me take you back through the three pieces one more time okay.

Okay.

On the -- to increase the growth rate assumption that we had on the -- sorry to reduce the growth rate assumption that we had on appreciation and customer assets, that would bring us down to the 7%.

That was in the range of 3 to 4 times the overall amount of the charge.

The charge was 44, take 3 or 4 times that amount, it was roughly that amount.

Okay.

Offsetting that, going the other direction was an increase in the average life that we assumed primarily on the fixed annuities.

Now to triangulate that one step further, of the $44 million in negative charge that we took, approximately $26 million of that was related to less costs that we will be deferring that we deferred in this quarter and will be deferring going forward.

So the answer is three or four times 44.

Three or four times 44.

Okay, that's what I was getting on the.

How about variable annuity dollars, assets?

The separate account assets were $21.1 billion which are mostly variable annuity assets.

Okay.

I thought that might include fixed but it doesn't?

Include what?

Fixed.

No.

The fixed assets are within the owned category.

All right.

Okay.

Thank you.

Michael DeCollin from Lehman Brothers is online with a question.

Wondering if you could offer some commentary on the charge card interest expense going forward.

It sort of bounced down and up in recent quarters.

I know in the past you have tended to lock in, you know, next year's rate today.

Have you done that?

And how should we think about this line item as a percentage cost going forward?

What we have you know tried to do obviously is in periods when we have declining interest rates we try and extend the maturities so that we can benefit from the lower rates going forward.

In periods of thriving rates, we do the opposite, we try to shorten.

So obviously, given the period that we've just been through we've tried to extend out a bit.

And so a big portion of next year's interest cost is locked at this point.

We have left some of that open you know in the hope, frankly, there would be continued benign interest rate environment which would be consistent with our overall views of the economy for next year.

That being said, it's not a big enough piece that it is likely to have a material impact on our results next year if there was you know deterioration in that number.

We might have a little bit of upside, but basically it should be a fairly neutral factor in our results for next year.

So would it be not inconsistent with third quarter levels or slightly better?

Well I don't think I should get into doing a point estimate for it.

But I think you assume it will be a neutral factor a full year basis for us.

Okay.

If I might ask one last question: To the extent that you guys have seen billed business trends in the month of October as we're near the end, could you comment on that at all?

I can't.

I'm sorry.

Okay.

Thanks.

Karen Mayer from Banc of America Securities is online with a question.

Please state your question.

Hi guys.

Just one quick follow-up question.

You gave in the release some sequential PNE spending trends.

Could you give us month-by-month consumer trends; and also could you give us overall billed business trends just to get a sense if, you know, absent the -- I know that there was a 9/11 impact, but it would be helpful to have the trends.

We don't split it out specifically, but let me give you a general sense, Karen, to at least be somewhat helpful.

Our consumer trends were pretty consistent in the first two months of the quarter with what we had seen last quarter.

And so then we saw obviously a big bounce in the month of September that then led to the domestic U.S. consumer billed business being up 10% overall.

On the corporate side, the corporate numbers as you recall last quarter were down a couple percentage points.

That's appropriate number to assume for the first couple of months of this quarter, and then we had a very large turn around in the month of September which gave us the plus 2% I believe in the corporate business.

That pattern is fairly similar for the small business business, which ended up 5% overall.

So the corporate side was kind of double-digits in the month of September to make up for the lower performance in the first two months of the quarter.

Thank you.

You bet.

I think we're getting down to maybe we'll take one more question here.

Bob Napoly from U.S.

Bancorp Piper Jaffrey is on line with a question.

Please state your question.

Thank you.

Question on expenses, just a quick follow up on the DAK The 26 million, is that the amount that you would expect to be higher on a quarterly basis going forward from prior levels?

That if we didn't do anything that would be the amount we would see full through on a quarterly basis.

Obviously, we'ring doing a lot to try to offset that, but has the amount we would see.

On the expense side, you guys have continued to recognize very strong improvement on expense line, more so showing up in human resource line;

I know some have switched to other from the IBM contract.

I was wondering if you could give more outlook, and has the majority of the benefit from the restructuring flowed through?

Or are there significant additional benefits to be received?

Or is it mostly incremental ?

Should we be seeing expenses grow from here?

The majority of the benefit of restructuring is fairly well complete.

You are going to see some benefit from that obvious l I in the first quarter but it basically has to be done by the first quarter of next year.

That being said, we've targeted I think we've reviewed this in terms of our long term thinking that we'd like to have our operating expenses if you exclude provision and interest cost our operating expenses trend towards the level we saw back in 1996 which is still you know 150 or so basis points away from where we are today.

And we like to get there you know over a three to four year kind of a time period that's imbedded in our long range financial planning assumption.

So we continue to have a very strong internal reengineering effort underway.

We've pointed to some of the major programs that we you know anticipate continuing to roll out such as our global infrastructure project which is moving some of our positions to lower wage environments.

The use the internet that kind of thing as a primary source of customer service.

A lot of the [INAUDIBLE] that we have going on from a procurement standpoint.

We continue to think we've got a lot of gas in that engine and our intention going forward is to try to improve this operating expense rate reflecting the reengineering opportunities we have.

The 150 basis points; is that a efficiency ratio as percentage of revenue.

Yes that expenses a percentage of rev new.

Just one follow-up.

The Lehman dividend, since that is disappearing and you've offset it with business build does that show up.

The business building or is it in the corporate line?

Yes it's in the corporate line as an offset to the Lehman dividends.

One thing nobody asked as a last question, if you wanted to talk about it; otherwise I can follow up online - The cash back program; how is that going and what are your goals for the growth of that program?

Well we have actually pretty aggressive goals.

We're pleased with the performance of the program so far.

We think it's absolutely terrific offerings.

We haven't given any specific details about the program yet, but we're pleased with what we have seen so far and we think customers like it.

It's good product.

What is the IRR in that product versus other products?

I don't think we'll give you that specifically tonight.

I can tell you though that as you might guess given our average company return on equity that all of these products have to return very attractive returns for us to make the decision to invest behind it.

And that product is similar to other products that we've had.

Okay.

Thank you.

Okay.

Thanks very much.

I think that wraps it up.

And we'll look forward to any follow-ups tomorrow.

Thank you.

This concludes today's teleconference.

Thank you for participating.

You may now disconnect.