ING Groep NV (ING) 2007 Q2 法說會逐字稿

Ladies and gentlemen, thank you for standing by. Welcome to the ING second-quarter results 2007 conference call on the 8th of August, 2007. Throughout today's recorded presentation, all participants will be in a listen-only mode. After the presentation, there will be an opportunity to ask questions. (OPERATOR INSTRUCTIONS).

I would now like to hand the conference over to Mr. [Thomas Wirheim]. Please go ahead, sir.

Good morning. This is Thomas Wirheim welcoming you ING's second-quarter 2007 conference call. Before handing this conference call over to Michel Tilmant, Chief Executive Officer of ING Groep; John Hele, Chief Financial Officer; Kos Timmermans, Chief Risk Officer; and Tom McInerney, Member of the Executive Board, let me first say that any forward-looking statements in today's comments are subject to a number of variables, including interest rates; foreign exchange rates; inflation rates; movements in securities markets, including equity markets; and underlying economic conditions and changes. The realization of forward-looking statements could be materially altered by unexpected movements in any or all of these and other variables.

Good afternoon, Michel, John and Kos. Michel, over to you.

Yes, thank you very much. We are here for about one hour, because after that some of us have to fly to London, and we will take the best time, [best of you], of this hour.

Given the fact that we have already made a few presentations today, so probably most of you have had the chance to look at it, I am not going to be very long here. I think that, for those of you who haven't heard anything yet, you know that we had a record quarter of net profits of EUR2.7 billion, up 36%. After the realized gain of about EUR573 million on ABN Amro shares, our underlying net profit went up 8.2%, and therefore a record quarter.

What is also very important is that after you have taken the noise of coming from the volatility due to IFRS accounting, it shows clearly that the business is doing very well. The performance of the business is up 25% in Insurance Europe, 30% in Insurance America, 20% in Insurance Asia-Pacific, 15% in Wholesale Banking, 14% in Retail Banking, and ING Direct had a fantastic commercial performance the last quarter with 500,000 new clients and EUR7 billion of mortgage.

So very good business quarter. I think we are pleased also this morning to report that we have limited exposure to the subprime mortgage, CDO, CLO and leveraged finance and that, based on [today market accountancies] we do not expect any material impact on 2007 earnings. At the same time, I think we continue to invest in the growth of our businesses, and you have seen in the recent past the acquisition of Oyak Bank in Turkey, the Latin American pension business of Santander, the South Korean fund manager in South Korea. So you have also probably read that we have decided to launch a retail bank in Ukraine, that we launched SPVA in Hungary and that we are planning an additional EUR65 million investment to boost growth of ING Direct in the second half of 2007.

I would like just to insist on the fact that those results are also the consequence of being very disciplined at ING, disciplined in strategy, disciplined in acquisitions, disciplined in execution, in capital allocation and also disciplined in risk management.

On this, I am ready to open the floor for questions.

(OPERATOR INSTRUCTIONS). Bruno Paulson. Please state your name and company name, followed by your question.

Bruno Paulson, Sanford Bernstein. A couple of questions. Firstly, focusing on the US Life business and particularly the new business profits, they seem a bit disappointing. If you break it down to variable annuity profits, new business profits look well down year on year. In Individual Life, the new business profit's running $9 million, which, if you take out the benefit [piece], the brought forth benefit from Q1 from the XXX financing means that Q2 on a standing-alone basis had negative new business profits. I wonder if you could discuss that and explained how it's going to be improved?

Secondly, on the whole subprime Alt-A issue, I was told you were going to give more detail, were willing to give more detail on the whole Alt-A exposure.

Also, I was wondering -- you mentioned the marked-to-market at the end of July. Things have deteriorated since then in the pricing in the subprime and CDO market, and I was wondering what the latest marked-to-market is in those areas.

Let me take the US Life value and IRR metrics. We have a new disclosure in the supplement -- I'm sure you seen it on page 9 -- where we are now breaking out for you the value of new business and IRRs for all the businesses. In terms of variable annuities, we actually had a good sales quarter at $1.915 billion, which was our best quarter in the last several quarters, I think probably a record quarter.

We did make a few enhancements to our products. Those enhancements have not fully taken hold in terms of sales, although we did have in the quarter -- June was a very strong month, and that's continuing. With scale of sales, it helps cover expenses. So we would expect the VNB to improve. In addition, because of the change in interest rates, with interest rates moving, the actual costs of the hedges were more expensive quarter over quarter in the second quarter.

So I would say on the VA, retirement services had a good VNB quarter. Variable annuity could be better, but I think we now have some good momentum in the product.

In terms of Life, that's a scale issue. Obviously, we got a benefit from the captives. But you are right; we are still not where we want to be on IRRs. Our strategy there, as we've said, is to increase our policy count, and we are going to get there significantly through our term sales. The last several years, our total policy count, Term Life as well as Universal and Variable Life, was in the 25,000, 30,000 range. We are now at a run rate to be at 100,000 policies. A lot of those are term policies, but there is a per-policy charge, so that covers a lot of overhead.

So we clearly need to make progress in the Life business. We need to get our sales up, and we need more scale in that business.

Sorry, just to clarify, you are saying that the target is a run rate of 100,000 policies. Is that per quarter or per year? You're saying you are there at the moment, or that's where you're trying to get to?

That is a run rate we are building to, and I believe, as of the last few months, we have been on that annual run rate.

So what kind of run rate do you need to get to, to get to a satisfactory IRR, given that the new business profits (multiple speakers)?

I think, in the 100,000 to 120,000 per year range.

Could you repeat your second question, because I didn't catch that (multiple speakers)?

I'm sorry, the second question was, I was wondering if you could give more -- there's been a concern this morning about the Alt-A mortgages, and I wondered if you could talk through the exposure there.

Secondly, the marked-to-market you gave, which was very welcome, at the end of July as opposed to the end the June, but looking at the prices, they have dropped quite a bit since then. I was wondering if you had any updates since the end of July on the subprime CDO marked-to-market, and whether you also had an Alt-A marked-to-market?

If I give you the Alt-A exposures, what we have [is it's grouped] in originated loans, where we have EUR1.7 billion and we have RMBSs. In the RMBSs it's EUR26.3 million. The RMBSs are split between ING Direct -- that's where the bulk is, the EUR24.4 million -- and the insurance company has EUR1.9 million. That is [a bit the] overview. So the bulk of what you see is in the ING Direct portfolio, which is for 99.9% the AAA-rated portfolio, what we have.

In a nutshell, if we talk about the marked-to-markets, no, I only have it for subprime and the CDOs, as per the 31st of July. I don't have further updates yet. But at the same time, by giving you the total size of the exposure, which is relatively small, you would see that the fluctuations I expect from that end not to be too spectacular.

If you talk about the Alt-A marked-to-markets, that is what I don't have. I [asked an] update over the July, and in general what you see over the RMBS portfolio, there is an offset in the development of the unrealized results, the development there. There is an offset between interest rates declining and credit spreads widening over the month of July, which was sort of offsetting. That's the only financial result which I have after the 30th of June which I can share here with you.

Richard Jenkins. (OPERATOR INSTRUCTIONS).

Richard Jenkins, Trimark Investments. Two questions -- can you just give us an update on your activities in India?

Secondly, a question regarding Taiwan, whether or not you would have been making any investments in fixed income overseas relative to the Taiwan book, and if that has been hedged back on the currency side?

Yes. Regarding India, as you know, we have three activities in India. The first activity is the Bank, and the second activity is insurance and third, asset management. I think, when it comes to the Bank, I think that you know that we have decided about a year ago to basically accelerate the rejuvenation of our Indian banking business. We have put a new management, and the first priority was to rebuild the infrastructure in terms of -- or build the infrastructure, in terms of people and management. As a matter of fact, in the last few months, we have been able to acquire very senior Indians to manage the Company. Second, to help and improve the overall infrastructure, and that is what we have done.

So I think, from that standpoint, the Bank has moved pretty well. I think that the strategic priorities for this year are clearly to increase our retail client base and to structurally improve the opposition of the retail savings by moving clients from term deposits to savings products and improve clients' service. So I think that that is -- the retail savings went up 6.3% in the second quarter of 2007, and the total number of retail clients increased about 25,000 in the second quarter to reach 1.23 million in June, so 1,230,000 people in June.

I think that we have reviewed the account opening process to facilitate and to improve the quality of client and reengineer it. So we have really improved our overall business.

As you know, the key thing in India for us is how quickly we can grow the business and the infrastructure, because the market is there. We basically say to our people, if you need more money, tell us. But at this point of time, they accelerate as much as they can. They opened four new branches in the second quarter. They add eight new ATM locations. The number of registered users of interim banking increased by 10,000 in the second quarter.

So I think we are getting some traction in India, and also the financials were better during the quarter. They are not yet significant at the ING level, but they are certainly improving and we are doing well.

In asset management, we have changed management and improved the infrastructure, and we hope that this will provide some benefits in the future. When it comes to ING Vysya Life, I think that what is important, that we now have 38,251 [tight] agents attached to 191 sales office in 173 cities nationwide, and our sales were up 39% compared to year to date through June 2006. So very strong increase in India.

The number of agents has increased significantly from last year, and therefore we believe very much that, again there, we are putting to our company there as much pressure to grow fast, and again there the issue is the ability to grow the infrastructure as fast as we can. We also have -- in the bank insurance business, we are trying to put together some bank insurance agreements, and we are successfully doing so with quite a number of [cooperative] banks. Any other questions?

There was a question on Taiwan, just if you were buying foreign (multiple speakers).

We do buy some foreign assets to lengthen duration, but we hedge the currency risk.

Dan Johnson. (OPERATOR INSTRUCTIONS).

It's Dan Johnson with Citadel Investment Group. A question on the US wealth management business. Can we talk specifically about the market conditions and the outlook for the 403(b) business in light of there's some regulatory changes underway in terms of both disclosure and fiduciary responsibilities for the plan sponsors? I would be interested to hear your views on that.

You are right, there's a lot going on in the market. I'd say, overall, in retirement services in the 401(k), (inaudible) the 401(k) and the 403(b) business, our core [K212] education business, have been going well. Today, in the 403(b) K212 education business, there are a number of companies that have agents within the school systems, and so that business has always been a one-on-one sale. The new 403(b) regulations in effect create the same type of marketing and distribution profile, where either a school district or a school system or an individual school becomes the plan sponsor. So what used to be a one-on-one sale by an agent to a teacher -- players now will have to make a sale to a plan sponsor, whether that's the school or school district or a school system, state system.

So we believe the large players who have been in the 457 business, [be that] the state and local governments and in the 401(k) -- who are used to making, in effect, two sales, one to the plan sponsor and then one to the individual -- will do well.

So we think the larger players who are in both 401(k) and 403(b) and the 457 business will do well. So we think those regulations will mean there will be a flight from a broad number of companies offering 403(b) annuities to the top three or five, and we're in that market. We have been with AIG and AXA in the top three positions. So I would think the top five, maybe top 10 players should do very well when the new regulations become effective early next year.

How about the regulations pertaining to fee disclosures within the both the 403(b) and the 401(k) markets?

Basically, we agree with that, I mean, the need for transparency, to disclose fees and breaking out fees between the plan provider and the asset manager. So we support that. We actually are moving forward with a one-page simple fee disclosure we are using in our business. I think you will find competitors following suit with that. We think that's good; it's clearly important for the plan sponsors to have that. There clearly will be more of a demand put on them going forward.

One of the challenges, I think, in the school district is whether it's a principal of a school or the superintendent of the town for the school really is an educator and not a financial plan fiduciary. So that will be a challenge, and one of the reasons that it has taken so long for these new regs to be put in place is I think there has been resistance from the school systems, the principals, the superintendents, to take on that fiduciary obligation.

But we are certainly for the changes. We think that's good for the individual teachers at the end, by having more of a plan sponsor focus to focus on the best plans, and also the disclosures allow both school systems to sponsor as well as the individuals, to make better decisions.

On the Japanese regulatory environment for your insurance products, can you just give us a sense as to where we feel that marketplace is sitting right now, relative to the regulatory environment?

The Japanese regulator takes quite a while to file your products there, as it took us last year. But we have very good relationships with them. It's not just the FSA; you also have to make sure your products meet the right tax regulations. Some companies have run afoul of that. We have been very careful with that, as well, and we checked our policies and our products with the tax authorities as well. So we find the relationship very solid, but they are very thorough, let's say, at their working through the products.

Well, I meant also in terms of the scrutiny around sales practices has just put quite a pall over the market in certain product lines over the last six to nine months, if you are aware of what I'm talking about.

Yes.

Is there any change in that?

Yes, especially in some of the more traditional products and some of the other sectors. But in terms of our products, which are sold through bank branches and everything else, we have solid disclosure and we have no issues on that front.

Paul Goodhind. (OPERATOR INSTRUCTIONS).

Paul Goodhind from Bear Stearns. I had a couple of questions, back on the famous subprime issue, if I can. The first is the marked-to-market figure you gave, I think, was EUR58 million. That relates to a portfolio of EUR3.2 billion. Does all that EUR58 million relate to the 2006 and 2007 vintages? You've given the breakdown of those vintages, which was quite helpful, but can I assume that, and that therefore there's nothing in terms of marked-to-market effects or previous vintages?

Secondly, even if that were the case, it seems a remarkably small figure. I'm wondering if you could just explain how that has been calculated. Because, not being expert on sub prime, I understand that there is little market, in many cases, or a not a liquid market. I'm wondering, therefore, how you have arrived at that figure, and whether you basically marked to model rather than marked to market. And if you have marked to model, have the model parameters been recalibrated to reflect the current market dislocation or not?

On the subprime part, the EUR58 million is the negative revaluation reserve over that total portfolio. I don't have the split-out for the vintages, but it's the total amount what we have there and what I know -- it's partly because that composition, I've seen somewhere; it's partly because of the higher rate of tranches and partly because of the lower rate. You have to keep in mind that the 95% is the higher rate and the 5% is the lower. But the total amount of negative marked-to-market is the EUR58 million. Just to be sure, that goes via our equity and only once we sell, which we don't necessarily plan to, it will be a P&L item.

If you ask, how do we calculate it, indeed, this is a bit of a challenge at the moment, because the markets are not all that liquid. Yes, we are following Lehman fixed-income indexes; that is one. The other one is the Interactive Data Corporation. So those are the two prime sources we are using for trying to sell you these portfolios, and we do it in a process which concerns both the front office of the business as well as risk management and finance.

[Kim Shapiro]. (OPERATOR INSTRUCTIONS).

Kim Shapiro at Citigroup. I've got two questions. The first is on the US Life Insurance business and profitability. Just wondering if the current level of profit in the quarter is a sustainable run rate going forward, wondering if we can expect further operating leverage effects as the account balances rise.

The second question is on non-Life profitability. This looked a bit weak in the second quarter, in terms of adverse claims development, so a bit weak adjusting for gains, hedge accounting issues, et cetera. Wondering how much of this is exceptional adverse claims development, and where we expect things to go from here.

Let me start with the US Life profitability. Obviously, if you look, it's probably easiest to focus on page seven of the US supplement, where in every single business line, the [Q7] was up substantially over the second quarter of 2006. So, really, we had a very good equity market, the equity market, the S&P 500 up 6% for the quarter. We assume in our pricing a 7.5% increase in the index for the full year, so obviously we would expect over time the index to be more in the 7.5% annual or whatever that is on a quarterly basis.

So clearly, that was an exceptional quarter. But there's no question, once the assets under management increase, our fees go up. So, as long as -- we will get the benefit going forward of where assets have moved up to, which we would expect. There was EUR33 million of DAC unlocking, based on the S&P being at 6% for the quarter versus 1.5%. So you wouldn't expect that every quarter.

We also had a very good quarter in terms of investment income. I think some of that is based on interest rates moving up. We should get some benefit going forward as spreads widen. While that causes you some negatives on subprimes and Alt-A's, you get the benefit in the traditional assets, which are where most of our assets are. So that ought to be helpful going forward.

We did have a very good quarter in terms of overall returns on our alternative investments, private equity. It was a good quarter. I think we'll have other good quarters, but I wouldn't say we'll have that level of private equity alternative performance every quarter.

Then the last area is in terms of our Life mortality results, we did have -- last year's second quarter, we actually had not a very good mortality, and this quarter we did. So there was a 22 million swing positive there. So over time, we would expect mortality to be closer to our assumptions. So it was a good quarter for Life, but I don't think we can count on that positive mortality each quarter.

So that covers the Life business, but obviously a very strong quarter across all the businesses -- wealth management and the insurance businesses as well as in the investment, asset management side. On non-Life, if you're speaking of the Americas, first, in Canada they have had a very strong positive result in 2004-2005. The cycle in Canada turned, obviously, as regulators sought to dampen premiums and premium increases. So the loss ratios have moved up, and we have disclosed that in the press release and the backup data.

I would say, relative to the industry, our Canadian business is number one in the market with an 11.5% market share. They have consistently beaten their competitors on average by about 500 basis points in terms of return on economic capital, and that continues. So we are at a place where the cycle is moving down, and that was the principal driver. There were some -- in past years, we have had some positive reserve developments on prior accident years, and we didn't have negative development, but we didn't have the same level of positive developments on prior years.

So overall, I would say on an absolute basis quarter over quarter, Canada is at a less positive point in the cycle. But on a relative basis to the industry, we continue to do quite well. Overall, the combined ratios are still well below 100.

In terms of Mexico, the challenge there -- in Latin America overall, the Life results were good. But in Mexico and our auto reserves, we do do reviews of all of our reserves from time to time. We did do a review in both the first and second quarter, and decided in the second quarter that we needed to increase the IBNR reserve for the auto reserves, and we did that to the tune of about 20 million.

Overall, in terms of our whole non-Life P&L line of business, which had an underlying profit before tax of 345 million, that had an allocation of the ABN gain from the corporate line of 74 million in those numbers; you can see that in the realized gains and fair-value changes in the quarter. The first quarter, non-Life, we had 229 million, so you can see how this is more in line with Q1. Clearly, we see continued rate pressure in The Netherlands non-Life, and we had a slight reserve strengthening in Belgium overall, but this, again, was benefited from an allocation from the corporate line.

Christopher Hitchings. (OPERATOR INSTRUCTIONS).

Chris Hitchings, KBW. Just three sort of technical questions have emerged from crunching the spreadsheets. One, just to be clear, you've made a net merger provision of -- this is Postbank/ING merger -- of EUR188 million. You have priced a EUR300 million pretax provision. Are those one and the same thing, or is there more to come in the rest of the year?

Second question is, yes, you've mentioned in the call this morning that you transferred a mortgage book from the insurance company to the Bank. I think you said it was EUR9 billion of mortgage books. That looks like all of the growth in the Retail Bank. Could you also talk through the profitability implications like [relevant] costs come across as well?

Third question -- at what stage will the Belgian business disappear from comparisons?

Fourthly, could you just talk us through the Australian Life sales? I think there was an accounting change at the start of the year which you didn't restate for. I'd love to know what the growth would have been without that.

If you [looked at] the first questions, about the provisions for combining Postbank and ING Bank, we refer in the press release of EUR188 million net. That corresponds to EUR252 million gross which were expensed in the first semester. That means that we have still to come, based on this, to about EUR48 million, for a total of EUR300 million of provisions [at this bond]. That's how you can reconcile that.

The second thing is, regarding the mortgage book, indeed, this transferred was of EUR9.4 billion. Now, what you've got to understand is that essentially the cost of maintaining a mortgage book are completely marginal. So the costs associated with that book are marginal, and I don't think that it's any material in any shape or form into the retail costs.

But was it most of the growth? Was it most of the growth of the Retail Bank? It would appear to be.

Yes, it is part of the growth of the Retail Bank in terms of revenue, and it will impact the revenues. But in terms of costs, fewer maintenance costs, and it's very, very small.

In terms of your question on Australia, we added, starting this year, a trust and long-term asset management business, the sales unit investment trusts, they are called there. It's now included in the figures; it contributed EUR70 million to the APE for the quarter, but only EUR4 million to the VNB.

So, in fact, it's most of the growth in Australia?

Oh, yes. You also had a question on the Belgian business. It will be shown -- once we've closed that transaction, the statements will be restated. (multiple speakers) sometime this year.

Sorry, just back to Australia, it was EUR70 million APE for the second quarter. What was it for the first quarter?

EUR50 million for the first quarter.

So, excluding that, Australia would have declined? If you had not made that change, Australia would have declined?

On an APE --

Sorry, have I got -- I'm muddling up second quarter and --

Yes. Second quarter, at EUR70 million, would have been up. It was EUR62 million a year ago, and would have been EUR158 million minus EUR70 million, so it would have been EUR80 million. Q1, I'd have to double-check on for you. No, we're still [up on] six months.

Farquhar Murray. (OPERATOR INSTRUCTIONS).

It's Farquhar Murray from Fox-Pitt Kelton here. Just three questions, if I may, mainly focused, unfortunately, on the subprime CDO issues. Firstly, just with regards to Alt-A, this may be very technical. But Could I just get a sense of your definition that you're applying there, just so I can compare it to others in the market?

Secondly, just with regards to the CDO exposures, could I just get a sense of whether that's a net figure or a gross figure, and what your hedge fund investments may have been doing in the CDO area as well?

Finally, just more generally, if we look at your exposures in both Alt-A and subprime, they do seem to be very much concentrated into the higher end of the high-grade structures. What I would interested in knowing is, really, has that always been the case? Or have you concentrated over time into that? If so, what drove you to do that?

First, Alt-A definition -- without getting too technical, if we talk about the originated loans in ING Direct, then we look at a few things, and that is the small portfolio -- or small portfolio, the EUR1.7 billion I mentioned. There, we are talking about the FICO score of 650, 700, LTV 70% to 100%. Please note that above 80% we will reinsure ourselves automatically at the same time, while there's also the case is we do quite some income verifications, especially if FICOs are below 700; as a matter of practice, we demand income verifications.

So that clarifies a bit the definition which we put ourselves around Alt-A for the originated mortgagees. If we talk about the CDOs, a good point we talked about EUR0.9 billion as a total portfolio. But that is indeed a netted portfolio, because in our financial markets business, what I have is I have a [loan bespoke] tranches of CDOs, where I buy protection in the index CDOs. So on a net basis, if I look at longs and shorts, then I have something like EUR2.2 billion of loans Longs against the EUR2.4 billion of shorts, which gives me a EUR236 million financial markets exposure in CDO, which comprises part of this EUR0.9 billion. Even there, my impression is that if you take that total amount of longs and shorts, what we do there is not an extremely large amount.

The last question was -- oh, the hedge funds. If you look at our hedge funds, there are two types of ways how we look at hedge funds. The first one -- let me give you a background there. In general, on the insurance side, we are not that high on the allocation of alternative assets and hedge funds, and that has to do with the fact that we were always a bit more long in the equity portfolio in The Netherlands, the 5% stake. So in general, we have a lower allocation towards the more alternative type of assets.

If you look, at the moment, at the hedge funds, then I know we have a total exposure of something like EUR500 million or EUR600 million, if I recall, on direct investments in the hedge funds. If you look at the type of hedge funds which we have -- it's EUR686 million, actually. If you look at the type of hedge funds, then it's basically recognized by the more multi-strategy type of fund, so there's an underlying diversification in the type of funds what we do, and it's not one single strategy what we choose there.

By the way, next to hedge fund exposure with direct investment, so that is what we have on the insurance, on the banking side we have our normal hedge fund financing or repo type of business, and there we normally -- well, we follow on a daily basis the kind of collateral what we are getting, and we on a daily basis also make sure that we do our margin calls on that. So that is, in a nutshell, what we have on the hedge fund exposure.

Then just more generally, on the third question, how has your your allocation in terms of AAA changed over time?

In terms of allocation, no, it did not change materially, although I do know that recent purchases which we did -- they are in the AAA type of category, but they are not a very high turnover in those portfolios. So if I look at over the month of July, it was more AAA tranches what was bought.

I'm looking more longer-term than that. Has your AAA changed over, say, the last three years?

Not that I'm -- no.

Trevor Kalcic. (OPERATOR INSTRUCTIONS).

It's Trevor Kalcic from ABN Amro. Just getting back to a quick question that Chris asked on the Dutch Retail business, the addition to the mortgage portfolio that's at roughly EUR10 billion additional mortgages -- could you just walk us through the P&L effect yet again on that? I don't really see that back in the revenue development for the Dutch Retail Bank.

Then the second question is, could you just explain to us why you actually did that?

Okay. The reason why we did that is essentially because it is -- first of all, mortgages are not an optimal instruments to have in the ALM of the insurance, number one. Number two, from a regulatory standpoint, it's better to have the mortgage book in the Bank and capital efficiency to have the mortgage book in the Bank at the insurance company. That's essentially what it is.

Now, I think that if we looked for -- compared to the first quarter 2007, the income increased by EUR8 million or 0.8%. Most of that increase was due to the transfer of the mortgage portfolio.

So, just to be clear, EUR10 million --

I think that if you want to have a ballpark cost of this, and if you want to put that in your model, put EUR2 million, and you get closer to what it is.

So, just to be clear, EUR10 billion in mortgages, essentially, that your marginal revenue increased per quarter, assuming that it happened in the beginning of [the old] quarter of EUR8 [billion], and that's pretty much it.

Yes, but it -- okay, it's a bit more complex than that, because the effective transfers was in May. I don't know the exact date in the month; it's at the beginning of the end. So there you're getting into real details that I don't have. But it's during the course of May that we transferred it.

Michael van Wegen. (OPERATOR INSTRUCTIONS).

Michael van Wegen, SocGen. Two questions -- first of all, getting back again on your Alt-A and the definition question there, you clearly highlighted in this call what the definition is of the originated book. This morning, you gave two numbers for the Alt-A RMBS exposure, being EUR25 billion, and then said, well, if we go to FICO scores of 650, 700, et cetera, it will be EUR2.3 billion. Can you give the difference in definition for the Alt-A of EUR25 billion and the EUR2.3 billion?

The second question is, to what extent is your investment policy for ING Direct in the US changing as a result of the subprime Alt-A issue we see currently, and what is the impact on your margins on the asset side and your growth strategy there?

Let me just reiterate, if we take -- what I gave you this call is on the originated side, I gave you these scores of what we do there right now. If we look at the RMBSs, yes, we just follow the widest definition in terms of portfolio of what we call Alt-A. So in general, we tend to call it Alt-A very quickly. At the same time, just to make sure overall, we have always been on the higher end of the spectrum in buying Alt-A, so always the very well-rated tranches. I can give you a bit more detail later on in precise granularity of what we mean with those definitions because, otherwise, it probably goes beyond the call to go into that in all details.

If you ask me (inaudible) do we want to change at this moment, the answer is no. It's 99.9% is AAA. At the same time, if we do witness some spread widening, we do not see that as something where we feel like we want to get out of that portfolio. We are more a buy-hold investor. So in that sense, it's maybe more an opportunity rather than a threat at this moment, in the very high-rated tranches. So no, we are not getting out of that type of business.

No, I wasn't suggesting you go out. I could actually imagine that you become a bit more aggressive, given that spreads are widening. You are more willing to give a better savings rate as well, and grow a bit quicker in the US, where you have been growing rather slow past couple of quarters.

I would say this is a good observation, but at this point of time I think we need to -- before changing this kind of policy, we have to think very hard on it. I think, at this point of time, the ALM committee has not, basically, approved any change in that policy. So it's too early to say.

There are no further questions at this time. Please continue with any further points you wish to raise.

Okay, thank you very much. Then I think that, if there is no more question, I thank you very much for your attention and your interest. Goodbye, and we close the call. Thank you very much. Bye-bye.

Ladies and gentlemen, this concludes the ING second-quarter results for 2007 conference call. Thank you for participating. You may now disconnect.