American Express Company (AXP) Q3 2012 Earnings Report, Transcript and Summary

Ladies and gentlemen, thank you for standing by, and welcome to the American Express Third Quarter 2012 Earnings Release. [Operator Instructions] And as a reminder, this conference is being recorded. I'll now turn the conference over to your host, Rick Petrino. Please go ahead, sir.

Thank you, Kathy. Welcome. We appreciate all of you joining us for today's discussion. 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, estimate, 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 and earnings supplement, which were filed in an 8-K report and in the company's 2011 10-K and Q1 and Q2 2012 10-Q reports, already on file with the Securities and Exchange Commission. The discussion today also contains certain non-GAAP financial measures. Information relating to comparable GAAP financial measures may be found in the third quarter 2012 earnings release, earnings supplement and presentation slides, as well as the earnings materials for prior periods that may be discussed, all of which are posted on our website at ir.americanexpress.com. We encourage you to review that information in conjunction with today's discussion. Today's discussion will begin with Dan Henry, Executive Vice President and Chief Financial Officer, who will review some key points related to the quarter's earnings through the series of slides including -- included with the earnings documents distributed and provide some brief summary comments. Once Dan completes his remarks, we will move to Q&A. With that, let me turn the discussion over to Dan.

Okay. Thanks, Rick. And I'll start on Slide 2, the summary of financial performance. So total revenues net of interest expense came in at $7.9 billion. That's an increase of 4% from a year ago. On an FX-adjusted basis, that's 5% growth. If we look back to the second quarter, reported revenue growth was 5%; FX-adjusted, 7%. So there's a slightly lower growth rate in revenues in the third quarter compared to the second quarter. Pretax income was $1.9 billion, 9% growth. Net income came in at $1.3 billion or 1% growth. So net income is growing at a slower pace than pretax income, and that's due to a lower tax rate in the third quarter of 2011. Diluted EPS came in at $1.09. That's an increase of 6%. So EPS is growing at a faster pace than net income due to our share buyback program. Return on equity is 26%, slightly above our target of 25%. And shares outstanding decreased by 4%, and this is also related to the share buyback program. So moving to Slide 3. These are the third quarter metrics. Billed business came in at $220 billion. That's 6% higher than the quarter a year ago, 8% on an FX-adjusted basis. If we compare that to the second quarter, in the second quarter, we had reported billed business growth of 7% and 9% on an FX-adjusted basis. So the third quarter billings growth rates were slightly below the second quarter by approximately 100 basis points, which seems consistent with the broader pattern of decline seen recently by others in the industry. Total cards in force grew 6%. That's a growth of 2% in our proprietary cards, which is consistent with what we've seen over the last several quarters, and a growth in GNS cards of 13%. Average basic cardmember spending grew at 4%, and this is reflective of strong customer engagement. Cardmember Loans came in at $61 billion. That's a growth of 6%. In the second quarter, we had growth of 4%. If we move over to Slide 4, so this is billed business growth by segment. And we see a slight decline across each segment in terms of growth rates. If you look at GCS, the green line, which is Global Corporate Services, it previously had a growth rate above the company average over the first half of 2012, and it has declined at a slightly faster growth rate to move more towards the industry average. And we've seen this slower growth rate primarily in T&E categories. If you look at Slide 5, this is billed business growth by region. And again, here we see a slight decline across each region, although a slightly faster decline in JAPA. And within this segment, we saw a slowdown in the growth rate in Australia, and that's having the largest impact on that region's growth. If we move to Slide 6, so this is a new slide. The bars are the dollar level of loans in each quarter, and the line is the growth rate in loans compared to the prior year quarter. So while Cardmember Loan growth rate has gradually increased over the past several quarters, loans at $61.8 billion this quarter are well below the peak levels of $77.1 billion that we saw in the fourth quarter of 2007. Loan growth of 6% in the second quarter compares to 8% growth in spending on lending products. If we move to Slide 7, so this is revenue performance. And if we look at total revenue growth, which is in the bottom right-hand corner, came in at 4%. That'll be 5% on an FX-adjusted basis. If we compare that to the second quarter, reported revenue growth was 5% or 7% on an FX-adjusted basis. So this quarter, we are growing slightly -- at a slightly slower pace than in the second quarter. Discount revenue is driven by a 6% increase in billed business offset by higher contra-revenues, primarily cash back rewards. I also note that the average discount rate decreased slightly to 2.53%, and that compares to 2.54% in the third quarter of 2011 and the second quarter of 2012. And as we've discussed before, it's being driven by pricing initiatives, mix and volume pricing. If we look at net card fees, increased 2%. And this growth matches the growth in cards in force. If we look at travel commissions and fees, it decreased 3% and is reflecting a 6% decline in worldwide travel sales, again reflecting lower T&E spending. Other commissions and fees declined 4% but was flat on an FX-adjusted basis. Other revenue is up 8%, and it primarily reflects a $30-million gain on the sale of ICBC shares. Net interest income growth was 6%, reflecting a 4% increase in average Cardmember Loans and slightly higher net interest yield. Moving to Slide 8, provision for losses. So the provision increased by $230 million. At a high level, we are seeing lower reserve releases partially offset by lower write-offs. So looking at Charge Card provision, it increased 9%. So this reflects a growth in receivables of 6% compared to last year. Credit metrics, as we'll see in a minute, are relatively stable. We had a reserve build in the third quarter of this year of $17 million, and that compares to a $27-million release last year. So modest changes in reserves. Cardmember Loan provision reflects loans growing 6% compared to last year. The delinquency rate is flat compared to the second quarter, and we continue to see declines in the write-off rate leading to a reserve release in the quarter. But as you can see on the slide, reserve releases this quarter of this year were $88 million compared to $421 million in the third quarter of 2011. So this has the effect of increasing provision year-over-year. And it was partially offset by lower write-offs. Write-off dollars in this quarter were $328 million, and that compares to $427 million in the third quarter of 2011. So provision is increasing while credit metrics are either stable or improving. If we move to Slide 9, so this is the Charge Card credit performance. And on the left, you can see the U.S. consumer charge write-off rate. They ticked up slightly in the first quarter of this year but have now ticked down slightly this quarter. I would view this as stable performance over the last several quarters. The right chart is international consumer and Global Corporate Services. As you can see on the chart, the net loss ratio is stable. And also note that these are historically low credit metrics. If we move to Slide 10, this is lending credit performance. And on the left-hand side, you can see that lending write-off rate continues to trend down in the quarter. Within the quarter, each month was basically stable at either 2% or 1.9%. Our view is that we are at or near a low point for the lending write-off rate. Our 30-day past due billings percentage trended down slightly in this month to 1.3%. And as I say each quarter, our objective is not to have the lowest possible write-off rate but to achieve the best economic gain where we make investments. These metrics are at historic lows and represent the best credit metrics in the industry. Moving to Slide 11. This is our lending reserve coverage. And so for both the U.S. and worldwide, reserves as a percentage of loans continue to come down as write-off rates and 30-day past due rates improve. Reserves as a percentage of past due have trended down as well for the same reason. Principal months coverage have ticked up a bit, primarily due to the low level of write-offs in the quarter. We think reserves are appropriately stated based on our credit reserve models. Looking at expenses, Slide 12. So if you look at total expenses at the bottom right, total expenses are 2% lower than in the third quarter of last year, and that'll be 1% lower on an FX-adjusted basis. Adjusted to exclude litigation settlement payments of $70 million in the third quarter of 2011, expenses would be down 3%. I will cover the individual lines on following slides, but I would note that our effective tax rate is at 33% this year compared to 28% in the third quarter of 2011, when we realized certain foreign tax credits. Moving to Slide 13. So marketing expense on Slide 12 indicated that marketing expense was $764 million in the third quarter of this year compared to $757 million in the third quarter of 2011. So that's only a 1% increase. So you could ask a question, is that a sufficient increase in marketing to drive our business growth? In fact, we have high levels of spending in both quarters. You can see marketing as a percentage of revenues in this quarter was 9.7%. We have indicated that we think marketing at 9% of revenues will enable us to drive business growth. So we are investing at healthy levels to drive future growth. If we move to Slide 14, so this is cardmember rewards expense. So cardmember rewards expense, as set forth on Slide 12, in the third quarter of this year was $1.496 billion. And that compares to $1.565 billion in the third quarter of 2011. So rewards expense is 4% lower than it was a year ago. As you can see from the chart, this is a combination of higher expense in the third quarter of this year related to higher MR rewards earned in the current period and higher co-brand expense -- that's the blue section of the bar -- but lower rewards expense related to the change in MR liability for points previously earned. So the increase in the URR in the third quarter of 2012 is lower than the increase in the URR in the third quarter of 2011. The net of these factors is lower rewards expense in this quarter compared to a year ago. The increase in the URR in the third quarter of this year is more in line with historic levels. The higher increase in URR in the third quarter of 2011 resulted from increased redemptions in that period. So let me remind you that for co-brand products, the co-brand partner has the obligation to deliver the reward. We pay the co-brand partner each month the amount we expense and have no balance sheet liability. On the other hand, we are responsible for delivering the rewards earned under the Membership Reward program and have a balance sheet reserve, which was approximately $5 billion at the end of 2011. So let's move to Slide 15, and let me remind you how the 2 elements of MR expense are calculated each quarter. So at the top, for points earned in the period, we took a look at total spending on products with the Membership Rewards feature, as well as any bonus points related to that spending, and we multiply it times the ultimate redemption rate and weighted average cost per point calculated for that quarter. The second piece is expense for points earned in previous periods. And when we have a change in the ultimate redemption rate or the weighted average cost per point, we apply those to all points outstanding at the beginning of the period. And that's the green portion of the bar. And it happened to be a small amount in the third quarter of 2012. Now while -- as we have said many times, rewards and loyalty programs continue to be a major competitive advantage for us. They drive billings, they have significant positive impacts on credit quality, and they result in closer, longer-lasting relationships with our cardmembers. We periodically evaluate the process for estimating the ultimate redemption rate and refine this process from time to time in response to changes in cardmember behavior and other factors. We currently have a review of our U.S. ultimate redemption rate estimation process underway. This review should be completed by the end of the year and will likely lead to a fourth quarter charge. As you know, our current ultimate redemption rate is 93%, a very high assumption for any consumer loyalty program. As with any such program, there are always going to be some breakage when participants leave the program without redeeming all the points they have earned. And as I said, the review of our ultimate redemption rate estimation process is not yet complete. I don't have any estimate to offer you today, but I did want to let you know that we have this process underway. Moving to Slide 16. So this is operating expense performance. And as you can see in the lower right-hand part of the chart, operating expense in the third quarter of this year is 2% lower than the third quarter of 2011. Adjusted to exclude the litigation settlement, it would be 4% lower than the third quarter of 2011. So this is clearly delivering on our objective of growing operating expense more slowly than revenues. Looking at salaries and benefits, it is lower by 5%. This is benefiting slightly from FX, and in the third quarter of 2011, included higher reengineering costs. Our employee count this year is consistent with our employee count in the third quarter of last year. You can see that professional services are flat. Occupancy and equipment is up 5%, reflecting higher data processing costs. And adjusted other is lower as a result of expense related to legal exposures booked in the third quarter of 2011. Now in this quarter, we established incremental reserves for customer refunds within adjusted other net, as well as several different revenue P&L line items. These reserves were not the primary driver of year-over-year variances in any of the single P&L line items that they hit. Earlier this month, the company announced that we had reached settlement with several regulatory agencies. Reserves were established in prior quarters for a substantial portion of these fines and estimated cardmember refunds. We are continuing our own internal review and also cooperating with regulators in their ongoing examination of add-on products in accordance with an industry-wide review. Let me move to Slide 17. So this is a new slide. We used it at the August Financial Community Meeting. So the green line is the growth in operating expense, the blue line is the growth in revenues, and the dotted green line is the growth in operating expense excluding litigation settlement payments. So let me make several points. The growth rates in 2010 and early 2011 were the result of our strategy to invest in business utilizing reserve releases and the settlement proceeds. In 2012, we stated our objective to grow operating expenses more slowly than revenue growth over the next 2 to 3 years. In the fourth quarter of '11 and the first and second quarter of 2012, excluding the settlement proceeds, we grew operating expense slower than revenues. This quarter, adjusted operating expense and reported operating expense are lower than they were in the third quarter of 2011 and are growing slower than revenues. This demonstrates our ability to effectively control operating expense. Moving to Slide 18. So this is expense as a percentage of revenues, and it shows adjusted expense. Adjusted expense excludes credit provision. On the left-hand side, we see 5 years of history, and on the right-hand side, the past 5 quarters. Over time, we expect this ratio to migrate back towards historical levels in 2 ways: first, through top line revenue growth; and second, through expense flexibility, which includes our plan to contain operating expense growth. Moving to Slide 18 (sic) [Slide 19], capital ratios. Tier 1 common ratio came in at 12.7% this quarter, which is similar to where we were in the second quarter of this year. We generated $1.4 billion in capital this quarter, $1.3 billion from net income plus $100 million from employee plans. We distributed $1.2 billion in capital, $1 billion in share buybacks and $200 million in dividends in the quarter. Risk-weighted assets increased somewhat due mostly to higher accounts receivable and loans. Our Tier 1 common ratio of 12.7% puts us in a strong capital position and well above required benchmarks. Moving to Slide 20, the total payout ratio. So this is a percentage of capital generated return to shareholders. The left side is the past 5 years, and the right side are the past 4 quarters. Our capital distribution plan allows for up to $4 billion in share repurchases in 2012. Over the first 3 quarters of 2012, we have repurchased $3 billion. We are maintaining very strong capital ratios while making these distributions. Moving to Slide 21. So this is a liquidity snapshot. We continue to hold excess cash and marketable securities to meet our next 12 months of funding maturities. We have $18 billion in excess cash and marketable securities, and the next 12 months of funding maturities is $16 billion. Moving to Slide 22. So this is our U.S. retail deposits by type. So we increased total deposits in the quarter by $1.2 billion up to $37 billion. Direct deposits increased by $1.3 billion, and third-party CDs decreased slightly. We remain committed to increasing direct deposits over time. So with that, let me conclude with a few comments. Given the uncertain environment, we feel positive about our financial performance, including our ability to continue to grow EPS in the third quarter in the absence of settlement payments and with significantly lower reserve releases. Spending growth continues to be healthy despite the uneven economy. Third quarter billings growth rate was slightly below the prior quarters, which seemed consistent with the broader pattern of decline seen recently by others in the industry. We also saw our average loans continue to grow modestly year-over-year, leading to 6% growth in net interest income. At the same time, lending loss rates improved to new all-time lows. Despite very strong credit performance, provision expense increased as lending reserve releases were significantly lower this year than last year. Revenue growth of 4% or 5% on an FX-adjusted basis slowed somewhat versus prior quarters and reflects the impact of a weaker economic environment. This stands in contrast to many other issuers who still face year-over-year declines. In the quarter, we demonstrated notable expense discipline with operating expense declining 2% versus the prior year. This is consistent with our plan of growing operating expense more slowly than revenues over the next 2 to 3 years. We are still investing in the business, and these investments are driving higher average spending and growth in the card base while continuing to build capabilities for the future. This was evidenced by marketing and promotion expense representing 9.7% of revenues this quarter, which is above our historic average of 9%. Our strong capital strength was also displayed this quarter as we were able to elevate our year-to-date payout ratio to 86% while maintaining very strong capital ratios. Looking ahead, we recognize that our business is not immune to the economic environment, but we continue to believe that our business model is well positioned for the challenges ahead. Thanks for listening, and we are now ready to take questions.

[Operator Instructions] And our first question will come from Ryan Nash with Goldman Sachs.

Just in terms of billed business, can you give us a sense of how it progressed throughout the quarter? I know that it was 7% at the Analyst Day on a days-adjusted basis, and it obviously came in a little bit stronger than that. And then second, just in terms of the European franchise, it was up 3%. In the quarter, can you give us a sense of how that looked by country? Did you see any weakness beyond Spain and Italy?

So as you said, we disclosed that on an FX and days mix adjusted. It was 7% in July. So we're up slightly from there. So obviously, the last 2 months were a little stronger than July. But I would say it was pretty even over those 2 remaining months. So there's not a trend going in either direction. In terms of Europe -- sorry, we had some background noise there. In terms of Europe, I think it's slightly slower growth than we saw last quarter. Notwithstanding that, it is down slightly but still at 3%. So again, we're seeing some slowdown, really, across Europe, so no major change in terms of how the various countries are performing on a relative basis to each other.

Okay. And then just in terms of -- can you just give us a little bit more color on the review of the IRR [ph], what drove it? What are the some of the key changes in your estimates that lead you to believe that there could be a charge taken?

Yes. The -- I think in the normal course, we periodically think it's appropriate to review the estimation process. It's generally driven by behavior changes by customers or as we collect more information. So in the first quarter of 2011, we made a change in the estimation process. Previous to that, we had only used attriter information, people who had left the program, to estimate the ultimate redemption rate for the active participants. At that juncture, we started to incorporate information from active participants. And so after that, we thought we would monitor it, and we thought it was appropriate at this juncture to do an evaluation of the estimation process.

Okay. And if I could just sneak one last question in, in terms of the add-on products, can you give us any sense of what percentage of revenue this is? And have you made any changes to the products that you're offering? Have you shut down any of the selling of any of these products?

So like everybody else in the industry, we know that regulators had a concern about them. So we stopped marketing earlier in the year, and we'll no longer offer Account Protector or ID Protect products as of December 31 of this year. And I'd point out that these products are not a significant source of revenue to us.

Your next question will come from Sanjay Sakhrani with KBW. Sanjay Sakhrani - Keefe, Bruyette, & Woods, Inc., Research Division: Back on that URR question, I was just wondering, have you guys considered possibly making changes to your programs, such that to influence the URR lower? I mean, or is it -- is there something that precludes you from doing that, like competition and market share loss? And then, secondarily, I was just wondering on the Walmart partnership and others. I was just wondering what the P&L implications are, just so -- I assume that there's some kind of revenue share agreement. I mean, do those get accounted for in the marketing line? And -- or are they contra-revenue items?

So Membership Rewards, we view as being a competitive advantage to us. As I said, they are an important part of driving billings. It also is a way for us to have a very close relationship to our customers. So it's a key element. It's not the only element. Certainly, cardmember benefits and superior servicing are all part of our value proposition, but Membership Rewards is a program that we think is the best in the industry. We actually think, to the extent we enhance the program and we have higher redemptions, that, that will have a very positive impact on the long-term health of our business. So at this juncture, we're not thinking of any initiatives -- wholesale initiatives to change the value proposition and drive the ultimate redemption rate down. If it continues to increase and drive billings in the future, then we would view that as a positive. In terms of Walmart -- so Walmart, we will, from that product, earn certain discount rate at the prepaid discount rate. We also will earn some float on the balances. There are no rewards costs related to this. There may be some minimal credit-type losses. There are actually no credit losses, but there could be some fraud losses. And really, what we need to do is grow this so that we can scale this and take advantage of what is a relatively low, fixed-base cost related to this product. So that's what I would say in terms of what you should expect in the future in terms of what lines it will hit.

We will go next to Mark DeVries with Barclays.

Dan, as we get closer to the next CCAR process after a year in which you didn't really deploy any significant capital on acquisitions, is there anything you can share with us about what type of payout ratio might be a reasonable request in light of your strong capital position?

I think last year, when we made our request, we wanted to remain [ph] flexibility so that if we were to do acquisitions, that we had built that into the submission that we had made, and also have the ability to distribute a significant portion of earnings if we didn't do acquisitions. And I would think we would structure our submission in a similar manner this year to give us flexibility in terms of what we do into 2013. And I think the fact that we have a strong capital base and the fact that we fared very well in the severe scenario that the Fed selected and had our capital drop to a much lower degree than most others, I think, are all contributing factors that put us in a position to actually make that type of submission.

Okay, got it. And another question. Your loan growth has really been consistently better than the rest of the industry over last year. To what do you kind of attribute that outsized growth relative to peers? And also, kind of what do you think the implications are for credit going forward?

I think we saw in -- throughout 2009 -- the flutter [ph] of 2009, '10 and '11 that we improved sooner on the credit side than most of the industry. And I think loan growth, as we were in 2011, was probably being impacted by the fact that people were deleveraging and we were seeing paydown rates increase over that period. But over the last 4 quarters or so, we've seen our paydown rates stabilize more. And as you know, our paydown rates are significantly higher than the competition. But the fact that those paydown rates are not going up, I think, is a contributing factor to the growth rate in loans gradually increasing over time. And as you know, we are very focused on Charge Card but also premium lending, and so I think all of those things are contributing factors to the growth we see in loans.

Okay. And any implications in there?

In credit? As you know, our credit metrics are at historical lows. As we think about where we go from here, we want to be thoughtful in terms of growing the business -- growing the premium business. Obviously, as you bring on new accounts, they have a somewhat higher risk profile just because they're at an earlier tenure in their life. And quite frankly, I've said this before, if you were to give me a choice of bringing in a group of customers that were going to have a 3% write-off rate, but had better economics compared to a group of customers that had a 2% write-off rate and lower economics, I would pick the group with the 3% write-off rate. Because again, we're not endeavoring just to have a low write-off rate, but good economic decisions when we do our investments.

Your next question is from Craig Maurer with CLSA. Craig J. Maurer - Credit Agricole Securities (USA) Inc., Research Division: I had a couple of questions. So first, back to the URR and just trying to continue to understand the trajectory here versus -- what are you finding when you look at -- you said you started looking at people that are staying in the program versus people that are leaving. How much higher is the redemption rate of those staying in the program versus those you found who have left? And secondly, if I could follow up on Loyalty Partner, we spent a lot of time discussing that when I visited you guys recently, and I know the growth in Germany and India is outstanding and you're coming to Mexico. But when you put that in a broader context of the additional $3 billion in fee income over 5 years that you had suggested, I think it was 3 Financial Community Meetings ago, where are we in that progress?

Okay. So in talking about the ultimate redemption rate, in the first quarter of 2011, we moved from using strictly data from people who had attrited to estimate the ultimate redemption rate to using that data plus information related to current participants, and we thought that would give us a better estimate. So that was kind of the change that we made at that juncture. The work we're doing now in terms of reviewing the estimation process is we're looking at whether we can enhance the segmentation of the information so that we can have a refined estimate that's enhanced. So that's the reason that we are looking at that. I fully recognize that saying that we're going to have a charge isn't particularly helpful to you in terms of understanding the amount. But at this juncture, we haven't evolved the models to a sufficient degree to have a reliable estimate. But I guess, in terms of trying to frame it for yourself, 2 things that you may think about is, first, as we disclosed in the annual report, if there was a 100-basis-point increase in the URR -- this is not a forecast. I'm simply quoting from the annual report. If there was a 100-basis-point increase, then, too, we would have a $330-million increase in the liability and there would be a charge to P&L in that period. I guess the other data point that I would give you, that we haven't disclosed before, is that if we had a 100-basis-point increase in the URR, that would have an impact of increasing annual expense by approximately $40 million. So that's in terms of just dimensionalizing it. Hopefully, it's helpful. Now in terms of your second question related to fee-based revenue progress, so we continue to expand the services we offer to cardmembers, merchants and other customers. There has been no change to the target that we put out there. We're about halfway through our time frame and still think our target is appropriate, although $3 billion is an ambitious target in an economy that remains so uneven. There's still a great deal of work to do, but we are moving forward on a number of fronts. Our emphasis will be on organic growth, but targeted acquisitions, such as Loyalty Partner and Accertify, may also play a role if we see the right opportunities. So hopefully, that answers your question.

Then we'll go next to James Freeman (sic) [James Friedman] with Susquehanna.

Could you share some observations or comments with regard to the fee-based revenue, if you might have an update in that regard?

So fee-based revenues is -- basically, the update is what I just said a moment ago. We continue to expand our services. We're about halfway through the time frame that we laid out to hit the $3-billion target. We recognize that it's a very ambitious target, given the economy. And we have great deal of work to do, but we're moving forward. I think we'll primarily get it through organic initiatives, but we may do targeted acquisitions as well.

So in that regard, Dan, are you still comfortable with the targets that Ken had set forward a year or so ago?

We think that target, which is to be at a $3-billion run rate as we exit 2014, continues to be an appropriate target.

Okay, one last one if I could sneak it in. So could you share some observations about APAC? I know it's not huge, but it was a little clunky. Were there any specific observations in markets like Australia?

Okay. So Asia Pacific, yes, so I think we -- so Australia is a big market for us within the Asian market and we have seen a decline in business there, particularly in the T&E segment. And we think that's probably a reflection of the fact that China is slowing down and Australia's economy has some pretty close linkages to China. So that's the biggest impact that we're seeing that is influencing the slowdown in the growth rate in billings in that region.

Our next question is from Bill Carcache with Nomura.

Dan, given the trajectory of loan growth and spending growth that you're seeing, can you just give some commentary around whether you expect spending growth and loan growth to cross, as you look ahead, such that basically loan growth overtakes spending growth? And then secondly, when does the loan growth that you're seeing lead to reserve building?

So I would not -- I don't want to forecast here, but I think loan growth is driven by 2 things: one, it's the growth rate in spending on lending products; and then the second aspect is customer behavior and whether they are still in a mode that they want to kind of de-lever, in which case, we're going to see higher paydown rates. Or at some juncture, should they move to a space where they're more comfortable and aren't focused on that, in which case, we could move to a spot where the growth in loans is very similar to the growth in spending on lending products. But there's lots of factors that play into that. Certainly, we have never ever had a target for loan growth. So loan growth is simply an outcome of the products we put out there that allow customers to revolve if they choose to and then how that customer utilizes the balance. But it emphasized that we are very focused on premium lending. We are not engaging in balance transfer, and we are looking to acquire customers that are higher spenders. We also have a focus on the fact that there are customers out there who are very good customers for us, are high spenders and carry balances at other institutions. And I think we'd be very interested in acquiring those balances. So certainly, just like Charge Card, premium lending has very good economics, and it will continue to be one of our focuses going forward.

And when does the loan growth that you're seeing lead to some reserve building that -- we've been seeing kind of releases overwhelm everything else, but is that something -- can you give us some color on when you expect that to change?

Yes. So I think, as we grow the business, we have always maintained the same types of credit requirements. Although certainly, as we look to get deeper penetration into premium lending, you bring on new customers. And any time you bring in a cohort of new customers, at least in the first 2 years of their life, they tend to have higher credit losses. So that could be one element of it. And again, we -- as I said before, we don't target loan write-off rates. We target good economics. Obviously, the percentage of people who are in that lower-tenured group could be a factor, then obviously the economy. If the economy continues to be strong as we grow our business, you'll probably see less of a movement. If we see some deterioration in the economy, that'll obviously influence it as we go forward.

Okay. And finally, if I can just do one last one, and going back to the fee-based questions, can you just clarify for us whether you consider bluebird, Serve and just prepaid in general to be part of your fee-based initiatives? And are revenues from prepaid, I guess, included in that fee-based revenue target that you put out there?

I think certain elements would be. So to the extent we get discount revenue on bluebird, you could look at that as it's a fee, because it is a fee. And certainly, other fee elements related to Serve would fit there. To the extent we have interest income on float, I wouldn't consider that to be a fee. So it may -- certain elements in the product will be fees and others will not. But certainly, they would be part of what we're looking to in terms of achieving our target over time.

Okay, and I'm sorry, but just to be clear on that, so the volume that you're going to get on Serve and bluebird, that's going to get thrown into discount revenues and that will drive fees, but the fees on that will count as part of the $3-billion fee-based revenue target.

Yes, it is. And I think it's logical because there's really no credit risk associated with those fees. So I kind of view it a little different than we might discount revenue related to either charge or lending products.

And we now have a question from David Hochstim with Buckingham Research.

Could you just provide a little more color on what you're seeing in the decline in T&E spending? Is it fewer transactions? Is it lower prices on hotels or airline tickets? I noticed there was an increase in airline spending, about 2%, I think, FX adjusted. But I'm wondering if there's economic slowdown occurring in more than just Australia.

Yes. So I would say that we are seeing lower T&E spending compared to other categories pretty broadly. It was a contributor to Australia. It also was a contributing factor to the slowdown in growth rate for Global Corporate Services. And I think we see large corporations, in particular, being a little bit conservative here and seeing a drop in spending among some of our larger corporate clients. So it's pretty -- I think it's pretty broad-based.

And spread geographically as well.

Spread geographically as well, yes, and in some lower levels of transactions. But notwithstanding that, when you consider all those negatives I just said in that sentence, we still are managing to have good business growth on an FX-adjusted basis of 8%.

Right, okay. And then I think you referenced higher cash rebates. Can you just give us an idea of how much those are as an offset to discount revenue line?

Well, it's not an item that we separately disclose. It is one of the items that represents the difference between the growth in billed business and the growth in discount revenues. And I guess it's a reflection of the fact that our cash-back products are being successful in the marketplace, so that pleases us in terms of that product category.

And then finally, could you just clarify what you were saying about the regulatory and litigation reserves? It sounds like you didn't take anything additional in third quarter. Is that to say that you don't -- and I guess, you don't anticipate, or it's not a problem, not estimable [ph], what's related to the protection...

So the majority of the payout we had under the regulatory order had been previously accrued. However, in this quarter, as we continued to do our own ongoing reviews, we did accrue amounts related to those reviews. So they were variable enough to mention but not so large that we would provide the dollar amount.

And I guess, could you give us a sense if they were more or less than a year ago in the third quarter? So was that...

So in the third quarter of last year -- I think in each of the quarters that we've had, I think they've been manageable numbers.

You have a big P&L, but just for the sake of...

That's a good thing.

Yes. But for the sake of kind of making sense of this quarter's expenses versus last quarter, can you just, since you don't want to tell us the amount, give us a sense of the size?

As I said, I think if you went back to the fourth quarter through now, there has been some amount in each quarter but none in any quarter that kind of crosses our threshold for disclosing the amount.

Your next question is from Don Fandetti with Citigroup.

It's been pretty quiet on the DOJ case in 2012. I was just curious, as you look out to '13, do you expect to see any procedural-type moves? Or would you expect it to be quiet again in '13?

I think what we're expecting is for discovery to be completed sometime early in 2013. And then after that, there'll be some time that elapses before we actually get to the next stage in the process.

Then we'll go next to Chris Brendler with Stifel. Christopher Brendler - Stifel, Nicolaus & Co., Inc., Research Division: Could you just give us -- sort of turning back to one of the earlier questions on the $3-billion target for fee-based revenues, I guess I'm not sure. At this point, it's been at least 1 year, almost 2, since you laid out those goals. I'm just wondering, what's going well? What's not? It seems like -- I would have thought we'd get a little more detail on Serve and how that's progressing. The bluebird product seems like a very compelling offer. It could be a big part of that pie should -- as you seek to achieve that $3-billion goal. Can you give us any detail on what the major components of that revenue target were and where you stand today in any way, shape or form?

So I -- there's certainly a broad set of initiatives. Certainly, Serve and prepaid reloadables would be elements where we would expect to contribute to our target. Loyalty Partner is an acquisition that we've discussed with you. We would think that's another notable piece of the pie. I would say, though, that if we were to sit here 3 years ago and think about the progress that would've been made in alternative payments and wallets and the like, you probably would've thought we were further along than we are today. So that's not a comment about American Express, specifically, but really about the industry. We continue to see a fair amount of press releases but not necessarily a fair amount of product that's in the marketplace. And the other thing, as I mentioned, is this is a pretty uneven economy. So it makes the achieving that target all the more challenging, but we do have a number of projects that are under way. And we continue to look at other initiatives as we go forward over the next 2 years, as we head towards the latter part of 2014. Christopher Brendler - Stifel, Nicolaus & Co., Inc., Research Division: Okay, a follow-up question on the credit card business. The lending margin was up this quarter. I didn't exactly -- understand exactly what was driving that. It wasn't like a huge increase but it looked, by my calculations, to be mostly on the top line yields, a nice little bump in the interest charges you're getting on your lending accounts. Anything driving that? I mean, your lending growth has been relatively impressive. I know it's a byproduct of your spending strategy. Just trying to think about what's causing the yield to go up at the same time the loan growth is picking up.

Yes, it's -- so as you say, it's only up slightly. It could be a little bit related to revolve rates but not a huge move in revolve rates. So we continue to watch it and make sure that the mix we have is right. Certainly, just a change in mix among products can also have a slight increase in terms of -- increase or decrease, depending on which way it moves. So we watch it closely. We've said that we want that number to be around 9% and that's what it's been largely over the last several quarters, slightly higher this quarter but not by a huge amount.

Then we'll go next to Ken Bruce with Bank of America Merrill Lynch.

Most of my questions have been answered, but maybe we could get you to dimensionalize some of the asset growth, the lending products growth. You kind of pointed out that spending growth is better than the lending growth and that there's a number of factors that are driving that. But could you, in any way, provide us with some sense as to whether this is an increase in traction for the premium lending strategy or if this is just the willingness of existing card members to increase their borrowing?

So it's hard to tease that out, to be quite frank. We are focused on premium lending. We have developed products to drive in that direction. As I said before, we recognize that there are some of our cardmembers who we know very well and we think have good credit who have balances elsewhere. That's certainly one of the things that we are thinking about. We're staying away from balance transfers and those kinds of initiatives, so we're not going there. And as I said before, we have never had a target for loan growth. It's an outcome of putting products in the marketplace as a design to drive spend and then the consumer deciding to utilize that. And how much they want to utilize that in part will be impacted by lots of things, including how they feel in terms of their confidence and where the economy is going. So there's lots of small moving parts within there. But there's not one overarching driving force that is impacting loan growth. And you could see that it's been a very gradual increase over time. And the fact that is the biggest is that 1 year ago or more, we were really seeing an increase in the paydown rate. And that was holding down loan growth effectively, and that has really -- it's increasing a little bit, but it's really stabilized over the last several quarters. So in terms of a change, I would say that's probably the largest contributing factor.

Okay, that's helpful. And then maybe if you could discuss in any way the decision to essentially rebrand Serve with a much bigger American Express prominence. I've found that to be quite interesting. I don't know if there's a thought that it just wasn't getting enough traction as an independent brand, and it just needs to effectively ride on the coattails of American Express. But if you could provide some sense as to what drove that decision.

So I think all new products evolve over time. And I think as a company, we came to the realization that our brand has a lot of attributes which are important to people who would use the Serve product or a reloadable prepaid. We do stand for trust, security and servicing. All those things are important when you think about a product like those. And so the decision was made to make it more prominent, so people who are purchasing those products recognize the linkage of Serve to the American Express family.

Okay. And is there any -- I mean, is it essentially targeting, from an aspirational brand perspective, a different demographic now? Or is it still the same and it's just a function of gaining more traction?

So I think -- obviously, people who are utilizing the reloadable prepaid, in particular, are a different group of customers than some of our premium products. But I think many of the attributes of our company and our brand are important to that customer group as well. And we think we have a platform in Serve, which is what reloadable prepaid is running on, and so we want to leverage that. And we think we can put products out there that can be very competitive in the marketplace.

That will come from Mike Taiano with Telsey Advisory Group.

So just wanted to follow up on the bluebird question. So it seems like the obvious rationale here is to push more volume onto Amex's network. But I guess, are there some secondary benefits as well? So in other words, from a data standpoint, do you benefit from getting access to a different demographic? And does that help you with some of your other businesses? And would it also potentially help you on the acceptance front as well?

So I'd say yes to all of those things, right? So I think it does push more volume onto our network. We think that is a very good thing. We think we can achieve good economics. We also see it as a large market today, I think over $300 billion. And it's forecasted to grow at, I think, 12% to 13% going forward. So it's a growing market. We have the capabilities and skills to be successful here. And so we viewed it as a good opportunity and also helping to grow to -- contribute to our growth in fee businesses.

Okay, great. And then just one quick follow-up on -- a lot of talk about the fiscal cliff here coming up and just wondering, does that affect your guys -- your planning at American Express in terms of the timing of marketing and your ability to maybe -- if there is a larger impact than expected, to ratchet back on maybe some of your marketing spend early next year?

So in our planning, we always have a base-case plan, and we have been scenarios. If things were to be better, we know exactly what we'd do. And if things were to be worse, we know exactly what we're going to do. So we will monitor this closely. But at the moment, we're focused on our base plan but do have these other scenarios in case we need to react quickly. Okay. So thanks, everybody, for joining the call, and have a good evening.

Thank you. Ladies and gentlemen, that does conclude our conference for today. Thank you for your participation and for using AT&T Executive Teleconference. You may now disconnect.