Well, thank you for joining us. We're very pleased we have Mark Lyons here from AIG. Mark is the CFO, and I guess that we're just talking still technically the Chief Actuary, but that will probably end pretty soon.
Hopefully.
I assume most people in the room know Mark. Before this, Mark was working at Arch Capital, both as the CFO and for a time, the head of the insurance practice there. I guess it's almost about a one-year anniversary of the move to AIG, maybe 11 months.
July 1st. Yeah, we're there.
July 1st. That's been very fruitful. Feel free, I'm going to open it up to the audience for questions. I'll try and get the ball rolling a little bit, but I'm sure that I won't be alone. Something on everyone's mind obviously, which I think people know, but we can reiterate a little bit on where the guidance is for 2019. Maybe what's going to be the thought about the practice of guidance after 2019, and has anything changed since December when that guidance was put in place?
Well, thanks Josh, and thanks for inviting me here today. Yeah, a few things I could reiterate. Josh, you remember from all my days, I hate guidance, but I love giving insight. I'll reaffirm some things. Brian had talked about the 2019 year coming in. There was a lot of unclear definitions, I think, of what it is. For the full year and the accident quarter and the accident year is sub-100, even with AAL, which I'm sure we'll get into. One, that was accomplished in the first quarter, and we continue to feel that will be accomplished throughout the year on the General Insurance side.
That's in terms of premium volume and investment income, I guess?
Yeah. Premium volume, we said would be roughly flat, and we continue to say that. Net earned premium last year, 2018, was $27 billion and change, and we continue to think it's the same thing.
With the premium volume, you've bought a lot of reinsurance this year to change the earnings profile, at least the volatility, and hopefully increase the earnings. Does that mean that on a gross written basis, AIG is now growing its business and able to offset the higher reinsurance purchases?
Well, you saw quarter-over-quarter on a GWP basis about 10%, 11% up quarter-over-quarter. A lot of that was due to Validus coming in, as well as more in Glatfelter. There were other areas as well. There were, I think to your point, you had that directional growth in gross, but in net it actually fell off a couple percentage points. Some of that was due to direct underwriting actions. Some of that was due to an increase in ceded. Definitely an increase in ceded. You got to also think about how does that stuff get booked? When you have quarter shares, of course, it kind of maybe lags 45 days, but it pretty much follows the gross. On XOLs, CATs, per risks, the booking, different companies do it different ways.
The way AIG does it is you get an annualized for a 1/1 contract, which is what most of them were. You get an annualized estimate up front. You book it as written, then it's earned over a full year term, as opposed to having four written estimates that are earned over a quarterly term. The earnings or ceded earnings are unaffected, but the written recognition is different. It's more front-ended.
The reinsurance program will certainly, I think, bring down catastrophe volatility. The hope is it's going to help contribute as well as underwriting, of course, to improve profitability. Is there a trade-off between the loss profile of the reinsurance transactions and the expense profile and how they affect AIG?
Well, normally, well, you buy it for a couple of reasons. In any market, in any company, I think AIG's was skewed more towards volatility containment, if I had to rank it first. Secondly, was to change the net mix of business towards your better business. Third is generally capital benefits or something. This was not done for capital purposes. I think what we'll see over time as the front-end gross underwriting, which any good company has to do, is fix and have strong gross front-end underwriting changes and our views of it changes. As to what we're changing internally, the pricing we're pushing in certainly selected markets, I think gets outside a corroboration. That will affect our view of what we do with reinsurance.
I would expect some of what we put in place we're probably going to renew, and some of it may be lessened a bit. There probably would start to be, outside of CAT, a more quota share usage than XOL usage.
I guess on the topic of being done for capital reasons, which you're not doing it for so much, where do you stand right now with the rating agencies and what their demands are and how that's directing your strategies for underwriting at this point?
We've gone through all of our meetings with all the four majors. I think Fitch already came out and reaffirmed. S&P took off. We're still A+, but took off the negative watch, which may have been a drag on the company. That's been restated to stable, so we're very happy about that. Moody's review, we haven't seen anything in writing. I expect that to be pretty much the same, but can't really speak for a rating agency. I think AM Best is still going through its analysis and deliberations, but we expect to hear from them in late June.
In terms of capital management, I think that, look, if we go back in time six months ago, the stock was below 40, now the stock's probably 52-ish. I don't know, we'll look today.
You have 36, but who's counting?
Yeah. If we think about that pair, if we six months ago it was gung ho, we got to buy back as many shares as we can given the price, but price is different now. What's sort of the calculation? Brian wasn't really willing to give anything away on the conference call about his buyback intentions. Although you have almost a $2 billion authorization sitting out there if you want to act. I guess between debt retirement, share repurchase, M&A, buffer capital, how should we think about what AIG is going to do with the cash that it's generating right now?
Well, you certainly laid out the landscape. That's your competing alternatives as to what to do. That answer will likely change every quarter. Well, let me answer it first indirectly by saying, when I look at our debt to capital, it's too high for me. I think some funds need to be diverted to reduce it. Not a panacea, not a crazy fix it in one year, two year type area, but when you're pushing 30 points, no matter what rating agency, because they'll have a little different definitional metrics, that to me is on the high end, and I want to make some progress on that. You can do that overtly in the numerator, or you can print black ink like we did last quarter and help it through retained earnings.
I think in 2019 it'll be denominator help, and I think go forward, except maybe some minor things. On a forward basis, I'd like to be more overt about it. I view that as a higher priority than it may have been viewed in past years. Share repurchases, as you said, it's attractive more so on an absolute basis. On relative basis, not so much so because we're at 52-ish in land. It's always discussed every quarter. All of this stuff is not inadvertent. There might be some aspects of buffer maintenance and so forth, and there's no shortage of investment bank glossy books coming in. I view that more that we still have Validus that we're integrating, and there's some aspects of it that I think we've integrated with people and talent and giving broader definitions to some people.
some in cat modeling and other strengths that they clearly had, we're using that. That really needs to be Once the deal's shut, the real work begins. It's not like the deal's over. I want to make sure that that is tight as well.
When you say the debt to capital is too high, now my memory on it, there's a lot of very long dated debt with 8% coupons out there. I think since you've been at AIG and probably been planning these sort of things, the tenure has fallen probably by about 75 basis points, I guess, over that period of time. The best laid plans of mice on how easy it is to tender for and retire 20, 30, 40-year debt with an 8% coupon in a 3% U.S. 10-year Treasury environment versus a 2.3% 10-year Treasury environment.
Yeah. It does take a little bit of alchemy. AIG has roughly $1.3 billion to $1.5 billion of debt over the next few calendar years coming due. I think when I look backwards, AIG has done a good job of refinancing, take advantage of interest rates and cycles. That doesn't help the debt ratios and the leverage. Helps the income statement aspect of it. There's opportunities that come up from time to time in blocks. There's opportunities that come up in one, I can just be overt and simply not issue new debt to replace expiring debt. You might have noticed we did a small amount of non-call five preferred, professional preferred. We're very light on that in the capital stack. You do get permanent capital, and certainly from Moody's, you get 50% Basket C equity credit for it.
I think there's more place for that in the capital stack as well. By shrinking classic debt over time, printing more profits, and expanding the deck to include more preferred, I think is directionally where we're going to go.
I think the warrants are now in the money.
That's right.
The ability to retire warrants, and this is too illiquid versus retiring shares, or is there any sort of thoughts on that part of the capital structure?
Haven't given a ton of attention at this time. You're right, you got to be careful as to whether you're in or out. I'm more focused on what I just mentioned.
Yep. All right. I want to open up the floor to questions. I have plenty of questions, but if people want to ask them, we'll allow them to. I can just keep going. See? No one, no questions. You're doing a great job, Mark. Okay. John has a question over here.
Hi. Can you just remind us the reserve difference between the block of business you reinsure with Berkshire, and what can you say about that?
The question is what can I tell you?
The reserve differences between the business that you reinsure to Berkshire. Apparently, they have a different reserve than you, and there's quite a substantial difference.
How do you respond to that?
Well, I can't completely speak for Berkshire. That never stopped me before. First off, I doubt you're going to find mirror images of any ceding in assuming company. You just never see it because it's not disclosed. Firstly. Secondly is they're in a radically different. Put it this way, if I had a 0.02 to 1 premium to surplus ratio, I would have more options as well. Any news that they find is going to be positive news, right? It can't go the other way. But other than that, you have to ask [Issig].
I know, look, I'm happy to have you here. You tend to be very good at answering the technical questions as well as the You always get those answers, the easier ones. I was looking through the financials, and last year, I think, GAAP 2017 versus 2018, the combined ratio or the loss ratios were generally flattish. There might be a little bit of difference, but ex CAT, basically more or less flattish. If I look at the statutory financials, I see huge improvements in lines like workers' compensation, other liability, commercial auto. We're talking about potentially, in some cases, 500, 600 basis points of improvement, in some cases, 1,000 basis points of improvement.
A, can you talk a little bit about if someone's running through your financials line by line in the stat filings, why the improvements are so sizable on the stat side and they're not coming through on the GAAP side? B, if there is something going on here, does that bode well for 2019 results? Are we seeing that somehow earlier than we should? Okay.
I'll call that part A through H of the SAT question. Well, first definitionally, because I had to get underneath this myself when I joined AIG. Financial supplement information, 10-K information is the whole enchilada, and therefore, there's no intercompany reinsurance issues that matter. Schedule P is really a reflection of the U.S. pool. There's reinsurance that comes in, and then the adjustments in the consolidated Schedule P, which I have to talk to about because it's almost unreadable. You need a magnifying glass to see this stuff. Takes out the major, but not all of the internal reinsurance arrangements. When you step back and think about it, you got a couple of things happening, and then I'll get into some of the line-by-line things.
By the way, you supply these adjustments to shareholders if they want to see them. You make them, they're easy to find.
They're on the website.
Even if the writing is small.
That's right. That is true. You can download it and blow it up as big as you like. You've got a couple of things happening. As the, I'll call it, the innate U.S. business is shrinking. I'll say the core, pre-Validus U.S. business in some areas is shrinking. You're having higher percentages in the internal quota shares that are coming in. Some of them are removed in those adjustments, but not all of those are removed. You got a little bit of that happening. Some of them, like if I talk to commercial auto, for example. Commercial auto, it is reduced. It's reduced to still an unacceptably high loss ratio. When I have to double-check whether it's a loss ratio or a combined ratio, that tells me something, right? The reason I'm bringing that up is to be brutal on this.
When your loss ratios are close to 80th or 100th percentile in the industry, it's pretty simple to institute some of the changes. If we were trying to go from 65th percentile to 50th or from 50th to top decile, that's a lot more microscope and thin sewing together. There are big changes that we can make that have big impacts. Only part of which would be felt in actually year 2018, which to your question, portends well into, I think, actually year 2019. Commercial auto, I would view that way. Depends on the other lines. What were the other lines?
Other liability seems to be a big one that has that trend.
Okay.
Even improvements in workers' comp. Other liability, commercial auto is by far and away, 2,500 basis points of improvement year-over-year.
Okay.
The question I would have is why on the GAAP triangles, when we don't have the same granularity, but also we have, I guess, more internal actuarial discipline on the GAAP financials.
Well-
Why don't we see those numbers come through in the same year they come through on stats?
Let's look at the relative proportions between it. AIG's net reserves, loss and loss expense reserves, after ADC and everything else, about $51 billion. The annual statement will be 35-ish, 33-ish. That's two-thirds. Two-thirds is represented there. That's the unadjusted numbers, right? Some of that then goes backwards because it's being undone, which will reduce that ratio even more, probably approaching 50%. Of the loss ratios you're quoting, like on that's about 50% of the total action. If you have, arithmetically, right? You got some improved, you got the total's the same, you got some that have worsened. You've got worsening in personal lines in a lot of areas, even CAT controlled. Not only in the U.S., I think we spoke to that on the call, but also internationally. You've had minor changes in other areas.
Whether it is energy-related businesses and things like that, you see some movements. Getting back to, I think, the core part of your question is because that's the U.S. pool with the bigger intercompany foreign business stripped out. It gets back to the U.S. business. Other liability also mixed its business from year to year between primary and excess, with excess decreasing at a quicker rate. The changes have started to come in, only partially of which, because it's actually a year or so it's earned premium, would've been felt in 2018. The changes in casualty are just enormous. To try to understand this, the issues that we encountered at AIG, it's not anything we'd never seen before on the management team. We've all seen it. The difference is they're deeper and there's more of them simultaneously.
That's why you need a team that has a lot of experience on seeing these kinds of things and ranking them and going through and getting a team. I like to say that every area of General Insurance was visited by medical practitioners. Some got RNs, some got good doctors, some areas got a team of surgeons, and that was North America Commercial. That got a team of surgeons. Everything got medical attention, but that required a lot more. Some of the changes I'm highlighting and you're observing, I think are completely reasonable because of where they were and the changes that have happened. The kinds of things we don't talk about, because it's not as quantifiable, is the shifts of distribution and welcoming in the wholesale channel, which really had not been fruitful or actually even been material, really, to AIG.
The way Lexington is being re-stood up, if you will, creates a wholesale strategy, opens up a whole channel that wasn't really used before, as well as a whole set of new wholesale producers. That moves you down scale also into the size of accounts that are there. AIG has a reputation of writing the big jumbos, and in fact, that's altering and that's changing. There's distribution changes, risk appetite changes, gross underwriting changes, reinsurance changes, all by themselves were pretty massive. As the funnel broadens and the risks coming in are more diverse from distribution channels, more tailored to the risk appetite that's been overly communicated on the outside world of distribution, and that's coming in, and even if you don't change anything on rate level, your rate adequacy increases because the average quality of the risk coming in is better.
Interestingly, you just gave me a stream of consciousness, I apologize. I'll go not too long here, when we measure the rate adequacy, everybody talks about renewal rate change, right? That's a standard measure, and we do too. When you look at how the portfolio is changing in any cluster of lines of business or sublines, we look at the relative rate adequacy of new business coming in, of lost business, and of the business that was renewed, as well as what were the changes on that renewed business. There's no requirement. This is the new AIG. There's no requirement for premium volume to grow. Margin's the driver, not premium volume. Even if premium shrunk, if operating income went up or underwriting gain or loss went up. First, that's what we're striving towards.
I would view 2019 as another year of surgery, of portfolio construction that's continuing to, I think, reap benefits. That will continue to reap benefits in 2019. It's going to continue to reap in 2020. I think I had an incomplete thought there. The rate adequacy difference of new business coming in versus what's being lost out the door is enormous. Not marginal, enormous. That's, to me, the implicit rate increase on the book of business.
That business going out the door, there is certainly talk about we are now in a micro hardening market. I'm a big skeptic that there's anything more than losses out there because it's not like there's some pricing improvements out there, and people say, "Hey, it's because AIG has got religion and they're putting less price pressure." This business that you're letting go isn't that attractive, I'm guessing. At least someone's going to write it, I suppose. How do you think about the business going out the door? At some price, it's good business. Maybe I've actually heard it sometimes at no price. There's certain business that's good at no price.
It's going somewhere. You don't hear big cries of big self-insured retentions or that there's an insurance crisis and nobody can find coverage. Clearly going somewhere. If I break that into two chunks, let's say large commercial risks and D&O risks. Large commercial, AIG was writing multi-billion dollar policies out there, sometimes multi-year. Big nets. I won't go through all that. You've heard Peter Zaffino talk about that. Huge nets, not big nets. When AIG either doesn't renew it or stays on it at a much smaller capacity, for grins, let's say it was $2 billion expiring and $100 million or $150 million renewed, brokers have to go out find a lot of markets, and they might need seven or eight to fill in that gap in capacity, which means they probably had to go to 25 to get it done.
Initially, we weren't really sure of how much broker receptivity we would have because they're really having to work a lot harder to get it done. We got a few phone calls on that regard. Ultimately, on the ones we've retained, when we have dropped our limits gargantuanly like that, our GWP was almost a negligible difference. That's one of 2 things. It's either we raised our price for that, or what we were charging in the past for the capacity was woefully inadequate. The answer is both on that side. On the D&O side, though, AIG has always been a top 2 or 3 primary D&O market, an 800-pound gorilla. We were out at Arch competing against it. We could see that they were influencing terms and conditions but not really price.
The 800-pound gorilla attitude is being brought up on the front end, price is being pushed materially. To your point, I would say AIG on the D&O front is leading the increases. I would say some in property. In some areas, we're benefiting from the actions of others. In most areas, I think we're being self-determined, and others are benefiting from us. Let me go back to the broker analogy, like I made on the property side. Brokers now find that their placements are getting a little easier now because with a stronger AIG primary in place, it's easier to complete the lead access and what follows. On D&O, there's higher switching costs than you might think. Everybody likes to talk about national account business, streamline deals, commercial auto, all the collateral, all TPAs. It's tough to switch.
In D&O, there's legal contractual stickiness. There's a lot of manuscripting going on. A lot of those bigger accounts anyway are very complex. Years and years have been brought in to integrate something attractive to the customer, attractive to the insurer. You don't want to throw that out and start over again. There's a lot of motivation to stay. It's actually been very heartening inside the company of being able to push like that, see that it's sticking, sister lines of businesses are looking at that, and we're telling them to look at that and see that they should be doing it, too, that they can move the market. AIG has shrunk, but it's still a force, and let's use that force.
Okay. Expense ratio, looking, you guys have a lot of businesses in international lines and some of your warranty lines that are high expense ratio businesses. If we look at U.S. commercial, it doesn't look that out of whack with the rest of the market, but there's clearly some extra expense in there. Although talking to people I know who work for AIG feels like the expense thumb screws are on pretty tightly. How much more can AIG do on the expense side? What's the timeline?
Well, there has been a lot of, I think, strength around cost containment and expense management. Preceding this, I actually looked back, Josh, at one of your notes that you said that turnaround part two, or when is this going to happen?
Yeah.
You did your CNA XL comparison. I may have gotten the article wrong.
That's sort of right, yeah.
You made the comment of CNA, I think, having 225 bps of GOE ratio improvement. AIG has been there. It's been 240, 250 for two quarters in a row. One quarter you say is a fluke. We were saying, at least since I got there in 3Q, that this was underway, but it wasn't self-evident yet. Became self-evident in the fourth quarter, I think became more sustainably self-evident in the first quarter. That's about the only comparison I would make with CNA because that was something you had pushed out. There is still, and we view a lot of opportunities that won't be cutting into bone and muscle. There is still a lot of process. There's a lot of manual process. There's a lot of checkers checking the checkers, which I think is a lot of opportunity for improvement and change.
Some of which is automation related, some of which is just post-SIFI, let's reevaluate how things are done in the company. There's going to be a real emphasis on that. What you squeeze out of that is not only cost but efficiency. I think a lot of what we've done to date has been cost first, efficiency second. I think we've been looking more towards efficiency first, cost second. I still think there's more there to improve on.
If nobody wants to ask any questions, I have a warning. We're going to go on to life insurance. You can stop me from asking life insurance questions.
Thanks, Josh. Just wondering around the competitive environment and what you're seeing. Lloyd's of London has scaled back a bit of volume over the last year or so.
Is that making it easier for you to pick up some business, or how are you seeing things?
I think it's easier for us to push the market. Lloyd's has been disciplined. You look out, you see trade press, and you do see Lloyd's and AIG in the same breath in a lot of those. I would think it's good to have the companion to help, a sizable companion, to push back on the marketplace more than it's an opportunity for us to get business.
I think I can't remember if it's you or Brian, but you've talked about the double-digit ROE within the next three years. Can you just talk about your confidence or visibility in terms of getting to that target and the key drivers behind it?
Sure. It's a good swan song question. Yeah, I'll reaffirm that too. Back to your original question of double digits in calendar 2021, three years out. Interestingly, that also contemplates having, I'll say, our legacy Fortitude at the current level of ownership. It doesn't contemplate getting any improvements or betterment than that.
Does say those assets would not be written down to make that?
That's right.
Right. Okay.
That's right. In order for that to happen, what really. Look at AIG big picture. You've got life and retirement, you got GOE, you've got this legacy melting ice cube, and you've got the corporate, which is an amalgamation of expense, corporate eliminations, interest expense. Call it's a drag on our way. That's how I view it. To hit 10, you got to have L and R and GI both north of 10 for it to happen, and that's part of the plan. I think you're focused probably mostly on GI. That's been the drag on the operation. It won't be linear, but it won't be back-ended. We go from 2019 to 2020 to 2021. It's not like we're saying it's going to marginally improve 50 basis points and then the balance improvement's going to be in 2021.
It won't be linear, but it'll be better than the first example. A lot of that is going to come from these massive underwriting changes that we're making. It's going to come from the increased GOE discipline that we talked about that has opportunities not just in North America, but elsewhere. If you look at the, I guess we can look at the year, but even the quarter-over-quarter on international, the improvement felt there was all expense ratio improvement. The loss ratio, and I had to get underneath this when I came here because competing mostly from a North American lens and say, "AIG's trying to get in low 20s? They're low 60s? How is that possible? That's what Travelers does." Travelers is a North American-centric organization. I look at it like this now.
North America is a 67%, international is a 57%, and if they were perfectly weighted, it would be 62%. We printed a 61.8%. It was weighted a little bit more towards international. Big picture, you've got a big slug that's been running a pretty consistent 55%-57% actually, and there have been some volatility in that that created some of those spikes. Working on that to reduce that volatility and that mix and the same continuing what we do in North America. The acquisition ratio is a little tougher because that's just a consequence of your mix of business. All of those are going to be worked on. The net is, the path is to have GI north of that, L and R north of that, so that it more than offsets the drag of the corporate operations, I'll call it that.
Life insurance, which I know is your specialty.
Love it. I love to be alive.
Net flows were positive in the annuity business in 1Q 2019. It's been a long time since that happened, I think about two and a half, three years. It was on the strength of fixed annuity sales. How sustainable is a positive net flow environment for AIG's retirement business? Especially given 10 years lower now, maybe there was a surge into fixed annuities in 1Q because of the 4Q sell-off and people got scared.
Is that really a leading indicator of something positive to come, should the 1Q flow results be taken with a grain of salt?
Interest rates clearly are something people are going to look at. They're also going to look at spreads and market volatility. Market volatility from a prevention from going elsewhere was helpful. I think the drag of that volatility in 4Q-1Q is still in the minds of investors. Short term, don't see that having that big of a negative impact. Historically, we've seen that when you have some pressure on inflow on fixed annuities, you generally have a slowdown, maybe not proportional, but a slowdown in withdrawals and other outflows on it, whether it's because the competing alternatives aren't that great either, or behavioral finance, people are afraid to move it. Any reduction in inflow is not immediately or correspondingly affected on withdrawals the other way. That's part of a buffer.
I think the thing to keep in mind on AIG's book of retirement business, especially in the IR section, as you mentioned, is it's very well distributed. It's about a third, a third, a third between fixed, index, and variable. When I looked across at the competitive landscape, Pacific Life has that profile, but they're half the size. Most other carriers are dominated or predominant in one of those three, and therefore, I think they're more susceptible. Our customers can choose a little bit more what it is that they want and not having one product peddled. I think that part of it, anyway, is going to be helpful.
On the group retirement side, it seems to me this is a business that benefits greatly from economies of scale. There's a lot of competition out there. Does there need to be this much competition? Will there be fewer competition? Is AIG a consolidator of this market? Given where AIG is currently playing in 401k, 457, 403, are there barriers or, I should say, moats that AIG can defend its position without getting larger?
TIAA and Fidelity are the 800-pound gorillas. They dominate the record-keeping space of that. I think they have about two-thirds of that market. With that, of course, you get ancillary business. There aren't a ton of small guys to really gobble up. There's probably going to be continuing some level of consolidation on that. You can look at the competitive space and the competitive space in which AIG operates. When you go back to your 401k, 403, and your 457s, that's really different ballgames, right? 401ks are corporate. For years, I don't know, five years anyway, the flows have been negative out the door. You're getting boomers and others. It's not in accumulation phase, it's in distribution phase, and that's expected to continue. Those that continue to march headlong and expand in the 401k space, God bless them.
403s is the not-for-profit space, which is where we've really been focusing on. Yes, we're in healthcare and some levels of governmental and schools, and there's different school subsegments and all that. That, on average, is a younger workforce, so you're more on average in accumulation more than distribution. 457s are, if you think of them as like top hat plans or deferred comp plans. They're some of the same spaces. Those would get more of the emphasis. AIG's strategy, from what I can have gleaned from all the strategy documents that I've seen, is to really attempt to be involved in the life cycle of a client.
Whether it's group and you're in there early on and you got a lot of 20s and 30-year-olds in there, and you make things digital, you make things more easy on a topic that you sell, not buy, basically, is helpful, so that when you get to rollover time, you're top of mind. Average age of rollover is about 55 years. Going out of that into the IRA market is, you can clearly see that demographic happening, and that's an area of emphasis as well. Net-net, I see you minimize the negatives of the flows by choosing where you're going to be. The more you stay in pure corporate 401(k)s, the flows are going to be outward and against you.
I guess going full circle back to a little guidance, I guess we're not going to get this guidance, but maybe I can get something. Cash flow. How should we think about how quickly you can monetize the DTA? If we can think we can figure out what AIG's cash flow is on their earnings, maybe can we talk about a little bit how much extra we can put on from the DTA and how quickly you can monetize that given new tax laws and whatnot, and maybe what discount rate internally AIG thinks about it is appropriate for valuing the DTA?
If you look at our 10-K, you're going to see $15 billion of net DTA. $4 billion of that roughly is probably what I call normal that everyone has. It's not unique to AIG. It's not a result of the financial crisis. When we report adjusted ROEs and ROEs on book value versus excluding AOCI versus AOCI excluding DTA as well, it's that really $10 billion that's subtracted out. That's the atypical. The balance would be what any company has. Timing difference between GAAP and tax and discounting reserves and all that kind of stuff that creates those timing differences. That $10 billion you should think of as being split seven and three. Seven of NOLs, three of foreign tax credits.
I think it's fair to say with the trouble GI has had, that it's been the Life and Retirement division that's been the consumer of it. The foreign tax credits largely expire in the next three years, and we expect those to be utilized. NOLs is much more a function of general insurance and general insurance is improvements. To Connie's question, if we're going to be adhering to that 10% ROE by 2021, GI is going to be throwing off a lot of income that should be cash protected by consumption of the DTA. That's what we're anticipating on it. I'm not going to use chapter and verse of it, but it hangs together that we expect that to be consumed. I think it goes through I think 2028 is the lion's share of it. It'll extend longer than that, but I think that's the lion's share.
We expect to increasingly consume it.
In a normalized year, I guess, not really guidance, but your mathematics on that, the foreign tax credits can be used by either departments, and so therefore you have profit either one that can consume that but the NOLs only the P&C business?
No, there's restrictions on, like a 35%, 65% restrictions on what can be consumed. There's a higher L&R consumption of FTCs than on NOLs. It will consume, but GI needs to step up to really eat them.
I guess the goal that Connie was talking about will be on adjusted ROE again, excluding DTA.
That is correct
excluding AOCI.
DTA, with consumption, will be a little less to subtract.
You also, I guess, have, not that you guys exclude that, but there is a couple billion dollars of equity in the deferred gain on the ADC that also doesn't earn a return, I guess.
That's true. In fact, that's what goes through the income statement. You see, $55, $60 million a quarter goes through on that recognition.
On that goes through.
On the recognition of the deferred gain.
Yeah, the recognition of the deferred gain.
Amortization, I should say, of the deferred gain.
That is right. There's an earnings element, although some people exclude that from GAAP and some people exclude that from adjusted. I guess there's some different accounting for that. There's really 10 seconds left. That's it. I mean, they gave everyone their chances. We're done. All right.
Yeah.
Well, thank you very much.
Thank you.
Enjoy the rest of the conference. Thank you to Mark Klein. Thank you to Liz Werner. We'll see you at some meetings.