ING Groep NV (ING) 2008 Q3 法說會逐字稿

(Audio joined in progress). All participants will be in a listen-only mode. After the presentation, there will be an opportunity to ask questions. (Operator Instructions)

I will now hand the conference over to [Gemma Bux].

This is Gemma Bux for ING Group, welcoming you to ING's third-quarter 2008 results conference call. Before turning this over to John Hele, Chief Financial Officer, Koos Timmermans, Chief Risk Officer, and Tom McInerney, member of the Executive Board responsible for Insurance Americas, let me first say that any forward-looking statements in today's comments are subject to a number of variables.

They include interest rates, foreign exchange rates, inflation rates, movements in securities markets, including equity markets, and underlying economic health and changes. The realization of forward-looking statements can be materially altered by unexpected movements in any or all of these and other variables.

That said, good morning here. Good afternoon in Holland, John, Koos, and Tom. Please, John, over to you.

Well, good afternoon to those in Europe and good morning to those in North America. This is John Hele, and I will do a very short presentation, because we did have a presentation this morning and the presentation is on the web for those of you that may have missed it. I will not go through it again.

But there was a question this morning, which we might as well answer from the get-go. We had -- on slide 60 of the announced presentation, there's a breakdown into buckets of Bucket one, Bucket 2, and Bucket 3 of our ING Direct and Insurance Americas Alt-A residential mortgage-backed securities. And the Bucket 3 is greater than two times coverage, and in prior quarters, we had broken that out 2 to 5 times coverage and greater than 5 times coverage. And they wanted the EUR15.6 billion broken down into those two buckets. The numbers are between 2 and 5 times is EUR5.2 billion, and greater than 5 times is EUR10.4 billion.

I will say that again. So the EUR15.6 billion is broken down between 2 and 5 times coverage at EUR5.2 billion and greater than 5 times is EUR10.4 billion. I hope that answers the question we had this morning on that.

And with that, we would just like to turn to questions. Thank you.

(Operator Instructions) Farquhar Murray.

Farquhar from Fox-Pitt here. Two sets of questions, if I may, just starting with the economic capital model. I wonder if you just could -- could you give us the impact of the changes you made there, both on the banking side and the insurance side, separately, if possible?

Also, I wonder if you could just give us some color and perhaps the numbers around the different provisions you have under Basel II, the provision number and the internal capital model number.

Then secondly, turning to the Alt-A, the way you indicate basically you take an impairment of EUR130 million versus what you feel is a credit loss of EUR40 million. Could you just give us some rationale behind the EUR40 million? Also, additionally, if it transpires you were right and you have overprovisioned, at what timeframe do you think you actually have the writeback there? Thanks.

Why don't we start with the last one first? We'll turn it to Koos, the EUR130 million versus the EUR40 million. And I can talk about the accounting -- speak about the credit side.

I think, Farquhar, indeed, if you look at the EUR40 million, at what is the expected credit loss, the way how it works is we model those bonds and we see whether the credit enhancement is enough. Now in a nutshell, what that means is we look at, first, all the current delinquencies and we make an assumption that most of that is going to go to default, and we will apply the loss-given defaults which you see, the LTVs, which we have a scheme which is based on both the vintages as well as the product types. And that is a bit in conformity of what you see in the presentation of what I have.

Secondly, after this initial flush of delinquencies, then we start to normalize our annual default rates, but taking into account factors such as [recost] risk. But you have to make some assumptions of a leveling back to normality, which is either called through an SDA curve or something else.

But, yes, you do a lot of bond testing, and also what you do take into consideration is your prepayment rates. And you know, in general, you set a CPR or an annual prepayment rate. We have set that, I think, at around 10% or so. So you make a series of assumptions, and based on that, you look at the cash flow shortfall, and based on that, you make your provision.

Then the fact that you now provision EUR130 million for something which is EUR40 million, is there a way how to make that back? There's a few things which play a role here. Yes, I mean, unfortunately, the way how it works with IFRS is that provisioning means writing it down to a market value, so that is what we have done until so far. If at a certain moment your cash flows are more than sufficient, a way to de-provision it -- I mean, I would have to look at the IFRS books on that and I --

Yes, I think what happens is once you are certain that the impairment charge you've taken is too large, you get to write it back up, but you do amortize it, I think, over the remaining life of the security. So it's sort of a general rewrite up over time, versus a massive hit in any one quarter. You take the expected improvement and that would be amortized over the remaining life of the security.

With regard to the second question about provisions, internal capital models --

I think if you look at the provisions, I think what you have is both the sum-up of the provisions which is related to our IBNR, and that is the one -- we call it the portfolio provision, which is on the cycle. So in other words, we look at the PDs, the probabilities of defaults and the loss-given defaults, and what you will see is a bit of a migration that your probability of default will go up and your loss-given default will go up. So in general, you expect your portfolio provision, which is approximately a third, that's an element which you expect a bit to increase.

If you then look at the second component, which is your specific provision on cases, that is a bit more, yes, sort of a surprise element, because it is right now not sort of clear evident in the portfolio that you can say one sector is already heavily represented in our sort of watch list or in our chamber where you do expect the biggest write-downs. So on that one, it is still, technically speaking, idiosyncratic risk.

Okay. Regarding your question on the EC models, actually the impact of the changes were in the AFR, not really in the C models. In the Bank, as we disclosed, it was deducting -- we used to deduct the difference between the Basel II calculations and our own calculations. But given that we felt this was -- the EC is our own best estimate, we took that out. That was a very minor adjustment to the AFR.

And on the Insurance, we moved to a Triple-A bond spread. We found that the difference -- actually, swaps used to be trading at a rate that was higher than the Triple-A bond spreads, but the Triple-A bond spreads have increased quite a bit in the covered bond spreads, so it better reflects the illiquidity of the portfolio.

We're actually now working with the Chief Risk Officer form for the Insurance group as well as others to try to introduce this type of methodology towards solvency, too. And that had a larger impact. In total, both of these, in terms of the AFR for the Group, it was about EUR9 billion improvement.

Maybe a final word on that. Because if you look at models, we talked about loan-loss provisioning, IFRS, IBNR, and that is based on the cycle. If we look at Basel capital models, it uses -- on the one hand, it's similar, because it uses same components, PD and [LGD]. But at the same time, what you do there is in Basel you use a downturn LGD, so a bit more negative number.

Then the third component is economic capital, internal. And for economic capital, the similarity is again you use the same PD and LGD, but at the same time, you take into consideration country and industry diversification as well. But the same building bricks, the same building components, are available in each of these processes -- provisioning, Basel, and economic element.

Just one quick thing. Firstly, it's actually Farquhar. And secondly, on the Alt-A portfolio that you've taken the impairments on, have you looked through to what the adjusted LTV is on that portfolio, given current market prices?

Could you say the last question again (multiple speakers)?

On the Alt-A portfolio that you have impaired, where you have taken the EUR130 million impairment and then you're giving the EUR40 million credit loss, have looked through to the adjusted LTV on that portfolio, just to make sure that your other calculations make sense?

What we do in this portfolio is I know that the loan-to-value is changing over time, because if the house prices deteriorate, your LTV changes as well. What we do in terms of making our credit calculations is we make an estimation of the longer-term cash flow projections, and what we do is we use sort of a loss-given default table on this. And that table over time gets a bit more negative, if you see house prices declining.

But it's a longer-term forward prediction projection. I think one of the bigger assumptions in there is that all your future annual defaults, what kind of an LGD table do you use with it? And we use the current type of LGD table. So we assume not a lot of recovery of house prices in there. We do expect some back to normality and recovery of the PDs, but we don't expect a lot of recovery of the house prices.

So I think the biggest component is not so much a 5% plus or minus right now in the current table. The biggest assumption is that in 2018 on a CDR -- or on a default we still expect the same type of recovery.

Okay, thanks.

Duncan Wessel.

JPMorgan. Just one quick question on Alt-A as well. Do you anticipate any impact on your Alt-A securities from any sort of loan modifications or -- I'm not sure how the market works, but it seems like there's political pressure to modify some of the mortgage levels in the US. And I was just wondering how that would work for ING and your securities and whether that impacted your assessment of those securities. Thank you.

I think right now -- and maybe, Tom, you can add something to that -- I hear some stories, but that also has to do with the fact that we have a new presidency and then there is some talk about support for some of the median incomes. But I think it is too early to see that anything has been crystallized out on that area, which you could say already like, well, there's an expectation on either a haircut on a loan from our side or whether it is part of additional loans injected out of the $700 billion TARP. I mean, I wouldn't know. Maybe Tom, you have any ideas?

I think on TARP, there's a lot of moving parts on that, as I think you are all aware. What started out as a program to buy pressurized assets has turned into a capital injection program, which right now is focused first on the top nine banks in the US and now a number of the regional banks. And in addition, there are also some of the US insurance companies are also beginning to put in requests for capital.

In terms of the actual TARP purchasing assets, they clearly are looking at the reverse auction mechanism, although that's still being explored. In addition, there is -- Treasury and the FDIC under Sheila Bair are looking at a program -- and maybe this is what you're getting at, Duncan -- where there would be encouragement through the Fed and the Treasury with the banks. And you're seeing this in some cases; I think JPMorgan and Citi have announced some of this.

But what the FDIC and the Treasury are looking at is restructuring the underlying mortgages. I mean, it's easier to do, of course, where you have a whole mortgage still held by the bank versus something that's been securitized, so that creates complexity. But one of the thoughts is to rewrite -- and I've heard numbers anywhere from 3 million to 4 million mortgages -- rewrite those based on the current market values of the home and the affordability to the occupier to try to keep people in the home. And then that the FDIC, Treasury or Fannie/Freddie Mac would provide a second mortgage guaranteed by the government to the banks, so that it would mean the impact on the banks or the holders of these securitized securities.

So that's in the works. I think the thought is that that could help get to a floor on prices more quickly. But that is very much a moving discussion. And as I say, most of the focus in the last few weeks has been more on the capital injections, but I think they will -- we will look at that. I think there's quite a bit of interest on the part of the incoming administration as well as the Congress in terms of beginning to look at the mortgage holder and the individual homeowner versus just the capital injection. So I think there's more to come on that, and it's a little hard to say what the outcome would be.

Duncan, though, to be clear, we did not take that into account when we set our provisions and calculated out potential impairments for this past quarter. I think we would wait until the program is done and proven working before we would adjust our LGDs or our PDs in any of these calculations.

(Operator Instructions) Bruno Paulson, Bernstein.

Thank you very much. Some various questions. Firstly, the fixed annuity sales are rising, but from a relatively low base, certainly compared with the variable annuity sales. I wonder if you have had any comments about the extent of ambition there.

Secondly, there's been very fast growth of real estate lending. I saw a 32% growth referred to, and really quite low loss rates at this point. I was wondering if you could discuss why you think growing real estate lending at that kind of speed at this point in the cycle makes sense, and why you are confident that the loan losses won't deteriorate too much.

Finally, there was a fairly low result for Central Europe. Even if you take out the EU 41 million of derivatives, it was roughly zero. Is this to the extent of, if you like, retooling the Turkish business? And is this the sort of level we should expect going forward? Thank you.

Why don't I do the last question first, is Central Europe banking, I'm assuming you're talking.

Yes.

Yes. We had costs of the rebranding in Q3. In one day, all the Oyak Bank branches became ING Bank, with orange pins and everything else, so that was a cost. We are also opening more smaller branches all over Turkey to expand the banking footprint there. So there were one-time costs spent that also impacted the Turkish result in the quarter. And that's the major difference there.

Why don't I have question number one, the fixed annuity sales, I'll turn over to Tom.

Yes, in terms of more traditional fixed annuity sales, they did go up in the third quarter. Obviously, it was probably later in the third quarter probably going into September, where clearly there was significant sentiment in the US and individuals more cautious. And I think you will see that continue and that you will see an increase in fixed annuity sales going forward in the industry overall.

And I think -- assuming reasonable spreads on fixed securities, I think it could be an opportunity for us as well as for the industry.

In terms of real estate lending, I'll ask our Chief Risk Officer to --

Yes, I think the one thing is 32% growth is what I don't see, I mean, on the real estate finance side. What we did see is in the first quarter we were at 33.2 and it moved in the second quarter to 35.6 billion.

Year-on-year, I think.

So maybe year-on-year, indeed, if you look at it from Q3 '07 (inaudible) to Q3 '08, it's a high number. But if you look actually in the year of '08 itself, the developments are not that big anymore. And that is -- so, it's basically 3Q/3Q.

What you actually see happening over this year, quarter-on-quarter, is not a lot of growth anymore. I mean, why not? It was still growing in the first and second quarter a bit, but that is also longer-term loans. They temper out gradually, because you have some commitments, some pipelines and things like that. But I would say at this moment, the loan production on this area is relatively low. And you know, I normally sit in the credit committees, and I must say that we see more credit revisions rather than new production at this moment, and that is more evidenced by the numbers right now.

But I must say the first and the second quarters, you see some growth, but it's more modestly. And right now, it's actually relatively low. Nevertheless, I must say if you look at the (inaudible) of the real estate finance business, that it is still relatively healthy type of business with good LTVs.

Thank you very much.

(Operator Instructions) A follow-up from Farquhar Murray.

Thanks again for giving us the opportunity to ask the question. Just on the Alt-A portfolio again, once these things move into impaired, what's the inclination [versus] to sell them? Particularly, how does the actual capital requirement develop at that point, or is there any affect there? Thanks.

It's an interesting point, because if you impair it right now, I mean, one of the dilemmas what you have is that at this moment, I say that the underlying credit performance is better than what a market price implicates. So that is not a good reason to sell.

Now let us take the moment of impairment, then what happens is, yes, we need to take a P&L hit on that respective bond. But at the same time, the issue is still there that the market price is lower than the ultimate credit losses which you expect to have. I mean, by the time that your view is different -- and as you say, like, hey, I'm going to either not be able to hold the bond because I have a cash flow problem, which we don't have, or by the time that your views on the credit changes and it warrants to sell it at a discount of whatever, 50%, for neg am or something, then at that time you would sell.

But I would say the moment of impairment is not so much an issue. It's more the underlying view on the credit losses versus the market price. That's the driver.

Under US GAAP, you can never write up again, even though things get better. So the incentive is clearly under a US GAAP filer to sell the thing out. With us, we have the option, if we think economically it's a good thing to do, we can do it or we actually write it up slowly over our book over time. So we have more flexibility on it.

And on the capital side, is there a capital incentive to sell at that point? (Multiple speakers)

No. I think -- well, we could look at the different perspectives. I mean, the one is we have the regulators, and the regulators are more of our view, that they say like a negative revaluation reserve in as far as it's an illiquidity spread and for us it's a buy-and-hold portfolio, they don't recognize it. If you talk about the analysts, well, this is why we have the dialogue with you, and this is why we give you all the numbers and also the stress tests showing something like what is a credit component and what is an illiquidity spread component. So that is where we try to continue to have you informed on it.

So I don't get the request from a regulatory perspective. And from an investor perspective, I mean, I think for us the best thing to do is to sit and wait with the portfolio, but to clearly communicate in line with the credit migration what we expect this portfolio to do.

I agree, actually. Probably the best thing is to wait. But what I'm trying to understand is is there a capital increase as a consequence of this (multiple speakers)?

No, there's no fundamental change in either our internal capital models nor I think under Basel II that treat the securities in any different form.

Okay, so there's no capital increment due to impairment that would force you to sell then?

No.

Okay, great. Thanks.

(Operator Instructions) Thank you. There are no further questions. Please continue with any further points you wish to raise.

Either our 200 or 300 pages of disclosure is getting so good you don't have questions, or it's so much you can't get through it all in the course of the day. But either way, we'd like very much to thank you for joining us throughout the course of the day, and we'll look forward to being in contact with you over time.

Again, if you have future questions, please feel free to contact our Investor Relations Department -- be happy to answer your questions. Thank you very much.

This concludes the ING third-quarter 2008 results conference call on November 12, 2008. Thank you for participating. You may now disconnect.