Swiss Re AG (SSREF) Management on Q3 2020 Results - Earnings Call Transcript
Swiss Re AG (OTCPK:SSREF) Q3 2020 Results Conference Call October 30, 2020 3:30 AM ET
Company Participants
John Dacey - Group Chief Financial Officer
Thomas Bohun - New Head of Investor Relations
Conference Call Participants
Andrew Ritchie - Autonomous
Kamran Hussain - RBC
Paris Hadjiantonis - BNP Paribas
Ivan Bokhmat - Barclays
Iain Pearce - Credit Suisse
Vinit Malhotra - Mediobanca
Michael Haid - Commerzbank
Thomas Fossard - HSBC
Edward Morris - JP Morgan
Darius Satkauskas - KBW
James Chuck - Citi
Operator
Good morning or good afternoon. Welcome to Swiss Re's Nine Months 2020 Results Conference Call. Please note that today's conference is being recorded.
At this time, I would like to turn the conference over to John Dacey, Group CFO. Please go ahead.
John Dacey
Thank you very much and good morning or food afternoon, everyone from me as well. I'm here with Thomas Bohun, our New Head of Investor Relations to talk you through the nine months, 2020 results.
Before we go to the Q&A, I'd just like to make a few quick remarks about the release we put out this morning. You've all seen that the Group reported a net loss of 691 million for the first nine months of 2020, which reflects a strong performance in the third quarter, a net income of 444 million. In addition to the COVID 19 reserves -- sorry, the additions to the COVID 19 reserves amounted to 428 million in the third quarter, bringing the total to nearly 3 billion for the nine month period.
I'd just reiterate that 67% of these reserves remain IBNRs. Excluding the impact of COVID-19 the group net income increased to 1.6 billion for the nine months, up from 1.3 billion over the prior year. PMCs ROE, excluding COVID-19 was 15.5% and the business remains on track for a 97 combined ratio in 2020. We're positive about the improving market conditions.
Life & Health ROE excluding the COVID-19 losses was 9.7% with a strong premium growth of 6% supported by large transactions. I know the course has turned around, continues to progress with an ex-COVID combined for the year of 96% and a normalized combined ratio, that's significantly better than the 105 estimate for 2020 that we provided at the beginning of the year.
In Life Capital, we successfully closed the ReAssure sale in July and Life Capital's paid up to the Group, a dividend of $1.5 billion. We continued to build out the open book business with year-to-date premium growth of almost 20%. We continue to navigate the financial market volatility with an ROI of 3.4%, and ongoing active portfolio management to maintain the quality of our credit portfolio in particular. The impairments for the third quarter were zero on that credit portfolio, the impairments year-to-date were only 27 million on the portfolio, that's an excess of $30 billion.
Swiss Re maintains a strong capital position with the SOC ratio of 223, as of the 1st of July. We noted that the ratio remained above the Group target level of 220 throughout the third quarter, and I'd simply reiterate the 220 is a target. It's not a limit for us.
And with that, I'll hand it over to Thomas to manage the Q&A.
Thomas Bohun
Thank you, John, and hello to you from my side as well. Before we start the Q&A, I'd like to remind you to please restrict yourselves to two questions each and then register again, should you have follow-up questions.
With that operator, could we please take the first question?
Question-and-Answer Session
Operator
The first question comes from the line of Andrew Ritchie with Autonomous. Please go ahead.
Andrew Ritchie
Hi. Two questions please. The first one John, I'm just interested in, if you could give us a framework for thinking about second wave type impacts from COVID from a loss point of view. Now I appreciate things like event cancellation is more a prolongation of the first weighted average. But maybe just a framework as to how you're thinking about second wave potential losses. I'm wondering, for example, does it matter that it could be defined as a second event? And obviously, I suppose you've got some exclusion in place that weren't in place before. So, just some sense for that would be useful. The second question is on life. I just want to get a sense is the event obviously been booking mortality charges as the notifications have come through. Are there any concerns about the outlook for morbidity arising from COVID and what it could mean for your long-term health effects and whether that would have an effect on your results? Thanks.
John Dacey
I think Andrew, the one I started on the Life & Health thing and provide a little bit of color there and try to address your specific question. What we've seen is a to start with our major exposure for mortality in our Life & Health portfolio is in the U.S. with a secondary exposure in the UK. The increases in cases and potential increases in -- and Europe will be unfortunate, but probably won't have a lot of effect on our overall COVID charges.
Back in the U.S., what we see is that in the second quarter, when we booked approximately 500 million of reserves, we expected a higher incidence of insured losses in the last month of the quarter in June than we actually experienced. And so we found ourselves with a little bit of a positive coming into the third quarter and retrospect. The third quarter itself, we've booked the losses that we saw come through in July and August, largely according to plan and have an IBNR effectively up for the month of September.
So, overall there is mortality rate are close to what we expected, we've got a new parameterization maybe that would help you and other analysts who would suggest that, at least in the U.S. with a per 100,000 excess deaths, we would expect about 200 million of pre-tax losses. And that seems to be a more precise estimate, given the evidence we've seen through the third quarter on validity.
On morbidity, we are concerned about long-term impacts of COVID-19 on health and the long holders already provided some indication that there may be subsequent health problems for people who have been infected. I think, at this point in time, it's just premature to be able to make any serious modeling of what that might mean for our portfolio. But it is an issue which we are looking to track closely and we've continued to access experts around the world to see what information, new information might be available for us to ascertain what additional challenges are going to be on the morbidity dimension as well as effects of long term mortality.
Your first question was related to the second wave. I think that the framework you might want comes a little bit from what we were able to or what we did book in the third quarter where you see a big migration away from the property losses related to business interruption, into some combination of event cancellation and potential credit and surety. So, the events will proceed or actually not proceeding as a result of the pandemic and those losses will continue through the fourth quarter with probably some carryover into 2021, I would guess at this point.
On credit and surety, we booked, in the third quarter between the P&C and CorSo another, nearly 50 million of losses. I think as you see more long-term economic damage and related defaults, you'll see more losses here. The wild card continues to be the interplay between national programs, especially in Europe which we're supporting trade credit specifically, and the assisting companies that are maybe otherwise near defaults. So, we'll see how that plays out.
Mortality, I think I've mentioned, in terms of the play, and then your, I think fundamental question is to what degree? We think a second wave will trigger additional losses in the property book through the business interruption. And there, I simply will observe the level of IBNR is currently up on our property books, PNC real world with all of the positions have a IBNR is of over 80%, 87% I think, is what we've disclosed, and it's unclear how these claims are coming in. We believe this is our exposure, but in many cases, we're still waiting for the presentation of the claims, the agreement of what the occurrence might've been to trigger it and the accumulation that that's appropriate for it.
So, absent a massive long-term lockdown across Western economies, I don't think you'll see a repeat of the kinds of numbers that we've put out in the second quarter. We anticipate some modest movements, what you saw in the third quarter for business interruption across P&C and CorSo was an additional, close to 50 million. And that's kind of the unfortunate muddling along that we would expect absent something more dramatic current.
Thomas Bohun
Thank you, Andrew. Could we have the next question, please?
Operator
The next question comes from the line of Kamran Hussain from RBC. Please go ahead.
Kamran Hussain
Hi, two questions. The first one is on CorSo. I guess underlying you talked about some 96% combined big improvement, I guess versus a half year indefinitely well, below your target for the year. Do you think this is just tailwinds from best to frequency or should we put in 96 and our numbers next year and then add pricing? So, that's kind of question one. And then the second question in P&C Re, I guess at the half year you had 1.5 billion of COVID claims. It's only 1.6 at the nine month stage. Was it a positive development? Or did you just kind of change your assumptions based on kind of how things are how pains are coming in? Thank you.
John Dacey
On CorSo, I'd say, yes, we do seem to have some tailwinds. It's actually somewhat challenging for us to understand that the nature of the relatively low level of manmade losses in particular in the quarter and something we saw to a lesser degree in the second quarter of this year, and maybe that these claims are late and being reported. It may be that the lower levels of economic activity in the spring and early summer simply have meant that there are fewer losses.
But I would strongly encourage you not to take a 96 as a starting point, we don't think that's a new normal, we set ourselves you take COVID out and you get to 96, but you make some adjustments for what has been a modestly positive prior year development, and a more normal level of manmade losses. And you still -- you're still well below the 105 guidance we gave you, but you're materially above the 96. So, I think that's the starting point on there.
With respect to the COVID losses for P&C Re, what I can say is Q3 was largely coherent with what we thought at midyear would be a play out of COVID losses. I think you're right. The P&C Re numbers are a bit light. But we didn't necessarily expect a lot to be coming there in the quarter, where we have, continued exposure, I believe, is on subsequent credit and surety losses, potentially something in some other lines, but the business interruption.
Now, again we believe strongly that the triggering event, were the governmental lockdowns that occurred in the second quarter of last of this year and we reserved aggressively related to that triggering event. And so if that's the occurrence, we're comfortable that in the third quarter, there really wasn't much other mechanism by which we found ourselves exposed.
Operator
The next question comes from line of Paris Hadjiantonis from BNP Paribas. Please go ahead.
Paris Hadjiantonis
Firstly, on capital and cover returns, obviously, you are going to what most likely you're going to have a net loss for the full year depending on how the last quarter develops. At the same time your SST is slightly above your targets. Can you basically explain to offshore explain the thinking behind dividend decision tree to this year? What can you essentially tell us about you still consider the dividend you're paying shareholders being one of the good Archer yourself the equity story, we've seen Q3 despite all the losses that we're seeing due to COVID this year? And then secondly, I mean on COVID from a SST perspective, I know there's an assumption in there regarding additional losses. Can you give us an idea where those additional losses are coming from, might my initial assumption is likely, but I don't know whether you have an assumption about P&C and CorSo in that as well? Thank you.
John Dacey
Sure. So, let me start with the second question, if you will. On the SST calculation, you're correct. We have a expectation of future losses coming from COVID, baked into there. As I mentioned, our third quarter total 428 was coherent with what we expected at mid-year with what a third quarter loss could look like. And so we've gone ahead and projected out for the full year, and then some cases into 2021.
I think that, the nature of those losses expected in the first case as you say, some increased continuing mortality related to the fact that, the pandemic is ongoing and people unfortunately are still dying of the virus. It involved other losses in P&C Re and corporate solutions. Clearly the events cancellations, we've got the mapped out quarter-by-quarter about and how we think that they are likely to play out.
We're not particularly optimistic on that front. And clearly the credit and surety losses, I think on property. As I mentioned in the quarter, we didn't see much in the way of triggers for new property losses for business interruption. We've got some model, but I think they're mostly what we would expect is that the IBNR that we've currently put up will migrate in case reserves when we get a clear indications of legitimate claims that we're responsible for.
So, that's the position I think you should think about for future COVID claims with respect to the modeling we've done ourselves on. Capital returns, I'd observe, we've got a fairly clear order of capital management. The first is to make sure that the Group remains adequately solvent and capable of running its business.
I think I can't imagine that we can check that box any harder than having worked through this year. The way we have and end up on July 1st with a SST of 223. Again, above the target, I think the second is to maintain and where appropriate increase the ordinary dividend, that I think remains an important part of our capital management and I'd simply reiterate that we've already accrued 50% of next year's dividend in the SST calculation as of the 1st of July.
The second part would be accrued during the second half of 2020. I think, we believe there are important growth opportunities for us. We've been pleasantly surprised on CorSo that the volumes have not contracted as much as you would have expected when you're pruning a third of the portfolio, thanks to the pricing increases, and there has been places where we've seen some opportunities to actually grow the lines of business, which we're comfortable with.
But premiums overall remain down year-on-year instead on P&C Re we're up. Net premiums are at 9% in the first nine months, we think the pressures in the market for continued price increases into the January when renewables are strong. And we will look to see how we can grow that business, but growing our business in an orderly manner does not conflict at all with our ability to pay dividend.
Operator
The next question comes from the line of Ivan Bokhmat from Barclays. Please go ahead.
Ivan Bokhmat
Good afternoon. Thank you very much. I've got two smaller questions. The first one, could you please talk about the reserve developments over the third quarter, maybe the underlying reserve strength? I understand that there have been some additions? Maybe you could talk separately about lines like specialty and separately about the U.S. casualty which we spend some time in the past? And the second question, it's on the investment results. Obviously, already some volatility in fourth quarter, just wondering if you could maybe outline what kind of hedging strategy you've had in place, should we expect the same successful navigating from that volatility as we've had, as we saw in the first half of the year? Thank you?
John Dacey
Sure. So, on the reserve development, what you saw in or what we experienced in the third quarter was a I'd say a modest negative prior development for P&C Re, which was dominated by our decision actually, after the close in the first part of October to adjust an existing reserve that we set up for a major aviation loss where there's new information came to the market to ourselves as well. We thought, we'd been conservative, but the major increase in the expected loss that was communicated to us had us push another fairly significant amount into that reserve.
That reserve dominated the negative for P&C Re. And if you take it out, I leave it with what I would call a red zero. In corporate solutions, we had a black zero on prior year development, a modest release. Overall for the group, if you take out the aviation increase, we're still at a black zero. So, to gives you a sense that there's just not been a lot of movements other than this one very specific issue, which I think you can probably piece together.
With respect to the investment results, during the course of the second quarter into the third quarter, we've removed a fair amount of the hedges that we had up or put up starting in February of this year. We still have part of the equity portfolio hedged. I think we've come down a little more on the credit side. Having watched what the fed's actions have met for investment grade credit. Pricing, I think, most important there is, the adjustments we made to the credit exposures and trading out actually starting in 2019.
But continuing in the first quarter and early second quarter of 2020 to remove ourselves from the most vulnerable sectors, and we remain very comfortable with that. So I think I already made on behalf of Guido Fürer and his team, the great result in the quarter of zero impairments on the credit book for the quarter. I think, we're continuing to feel fairly comfortable about our positioning. There's obviously volatility in the market that I would expect will continue through next week. Let's hope it settles down, but if it doesn't settle down, we're still I think reasonably cautious in the positioning that we've got for our investment portfolio.
Thomas Bohun
Think Evan. Could we have the next question, please?
Operator
The next question comes from the line of Iain Pearce from Credit Suisse. Please go ahead.
Iain Pearce
Hi. Thanks for taking my questions. Two from me. Just trying to understand on the assumptions around second lockdowns that you made in the SST, and if there is an expectation of second lock downs that you baked into the SST number you've reported. And then on that as well, just in terms of geographical exposures between sort of France, Germany, UK where you see the biggest risks to your book around second lockdowns? And then second one on COVID losses. I noticed that, the credit and surety loss and the other loss in Q3 actually decreased in P&C Re. So, I'm just trying to understand what's happened there please.
John Dacey
On the second one, the decreases that you referenced, I think were fairly trivial in amounts maybe, less than 10 million and the positions and there's nothing really to say there, other than just, maybe some true-ups of modeling. But I wouldn't read much into it other than we were very close and there are some people that Swiss Re that like precision and might make an adjustment of 8 million along the way.
On the second lockdown, that's been modeled. We did not assume the kinds of lockdowns that we saw in March and April and to recur. What we expected was some partial lockdowns in certain regions, which would affect economic activity, but not wholesale. We do said, I think with geographic exposures, we see, what we've seen to-date is that, in the United States, our exposures are limited in part by the relatively high attachment points for the excess of loss treaties that we've got in place.
And by the fact that, the legal system seems to be maintaining a fairly literal interpretation of wording on the primary insurers contracts, and so there, the exposure seems to be contained, I think what we saw in the UK Court's FCA rulings was a, as I mentioned, a bit of a mixed bag that we anticipated. I don't think we're surprised even if it was maybe marginally more in favor of the insured than the insurers, other than we might've modeled. But we remained fairly comfortable. We didn't expect this to be clean in the following quarters. And so, our mid-term estimate suggests that there will be some additional losses along the way, but not a wholesale lockdown.
So, if that's the case, we're going to have to go back and evaluate what those exposures are a second lockdown at the same duration, same scale, same scope, is likely to have a less costly charge to us in part because of business that has renewed in April and in June and July. Obviously, the January one renewals are coming forward. And again, they're the contractual wording you should expect to be even tighter than what might have been existing with respect to exposure to pandemics.
Operator
The next question comes from the line of Vinit Malhotra from Mediobanca. Please go ahead.
Vinit Malhotra
Yes, good afternoon, John. Just two sort of follow-ups. First one is on the U.S. mortality. So, I mean, we're trying to understand how worse it has become in the last few weeks, even of the on the excess mortality side CDC data, I could check ran up to only 17th October. I mean, how are you thinking about this risk for your numbers? Because at least up to 17th of October, it didn't look like excess mortality was very high. It might have worked worse. I mean, are you how do you reserve the latest weight for your students to come back to you? Or do you make some estimate? And are you able to share whether this question was raised when you are booking the 3Q numbers? And I understand it's not booked. But it could just help us understand that because this question keeps coming up. And second topic is just on corporate solutions in the past. There have been some claims about PG&E, and you may not want to comment with one case or the other, but was not just curiosity, is that instrumental in some of the reserve movements that could have happened in corporate solutions? Thank you.
John Dacey
Sure. So, on U.S. mortality benefit, you're exactly right, that we're doing our reserving based on all cause mortality, actual versus expected. So, we're not dependent upon the classification of death has being a COVID death or non I think that was especially important. In the spring where systems were being overwhelmed and people were simply being or the people didn't have the time to clarify the exact cause of death. What we done through the third quarter is to estimate the number of deaths that we expect, actually occurred as of September 30th. And so subsequent to that, what what's gone on in October and you're exactly right.
There's a lag on the CDC data. And obviously, there's a lag on the information we receive from our primary companies, Although to be fair, they have been on this particular topic fairly responsive and forwarding information they have especially on the large dollar in short lives, that that might be relevant for us. I think we just need to continue to monitor it's a tragedy that in the U.S., we're back up to 1,000 deaths per day for the last three days and if this continues, and October in November in December, and the rough calculation with I told you, another 200 million per a hundred thousand deaths. It would be a relevant calculation for the quarter. We'll see how it plays out.
Our hope is that, some of the actions taken in individual states will improve. It's clear that, the mortality at those moments is very different than it was in the spring wave, protocols, procedures for treating patients is obviously materially improved at the moment. Health systems are not being overwhelmed with the way that either New York or Northern Italian hospitals, might've been back in April.
On the other hand, you see stories about some of the rural hospitals in the U.S. some of the hospitals in urban areas and France and Spain being close to full capacity. So, we'll see how this plays out. But we're a bit optimistic that actual mortality will be contained, and we just hope on multiple dimensions that the pandemic itself comes back under control in these jurisdictions.
On the corporate solutions the PG&E, we don't normally specify the source, but I think, the California settlement probably was not particularly relevant for our CorSo business in terms of the positive are coming through. I think what we're seeing is that a series of smaller positions, which had been trued up for prior years.
Thomas Bohun
Thank you, Vinit. Could we have the next question, please?
Operator
Your next question comes from the line of Michael Haid from Commerzbank Please go ahead.
Michael Haid
Thank you very much. Good afternoon to everyone. Two questions. First, also on the solvency test, the 223 slightly below my expectations, can you give us some moving parts there? In particular, I'm also interested in, did you assume higher growth in 2021, higher price increases? I think you already commented on increased COVID-19 loss expectations in your SST ratio. The second question, more a philosophical one that actually. COVID-19 and the problem of ensuring earnings, how is this for the new business? If the COVID-19 crisis develops, well into the 2001 21 year, then you might find yourself in the situation of insuring burning houses, maybe in facultative insurance. You already encountered this problem that, for instance, business interruption or event cancellation, but the likelihood of a cancellation is extremely high. So, how do you deal with this, that you have long-term relationships in place, on the other hand, you probably, already you ensure losses which have already, which you expect to come actually, how do you handle this problem?
John Dacey
So, with respect to the first question on the SST, I recognize that the consensus might've been a few points higher. I think, there were some modest adjustments to our own modeling, some foreign exchange movements and maybe the impact of some of the financial market hedges, which might have not been transparent to people and what's come through. I think, overall, 4 percentage point difference between where we landed and consensus should not be a big fixation for the market.
What I think we should be reminded is just the overall level of the capital strength, whether it's 227, or 223, or 220. Frankly, it remains one of the most robust capital positions in the industry. And in addition to that, I'd say, just reiterate that we face no red line at 220, that's a target we expect to be around that over a longer period of time. But dipping below that is not a particular cause for concern. So, long as we expect to, continue to move, plus or minus that number. So, that's the one. Your first question, I think.
On the second one, a couple of thoughts. One is on event cancellation, I think the industry broadly and I can say Swiss Re specifically, probably missed the what in hindsight is obvious accumulation effect of a pandemic on the potential set of losses coming through on a worldwide basis for these events that are insured, I think it's linked to the fact that while the pandemic itself is not a black swan event, the simultaneous lockdown of economies for 6 to 12-weeks around the world was not something that anybody had modeled.
And as a result, the subsequent play out of this probably makes event cancellation insurance, a harder cover to, to obtain, than previously, we've been very clear that we decided in the restructuring of corporate solutions in 2019, to exit this line of business. And so the losses you see coming through, especially through CorSo are effectively the existing book running-off. And to the degree that people write new event insurance for 2021. My guess is they'll be cautious about exactly what gets covered.
There'll be rethinking the required pricing. And in some cases, my guess is it's simply won't be picked up by the holders of those events, which is unfortunate, because it may create some bigger challenges to actually hosting events. But be that is a May, I think we're not. We don't expect to be particularly exposed there. I think more broadly on the business interruption. Source of losses, yes, the changes in contract, wording between reinsurers in primary companies between primary companies in their insurers I think will be much clearer about exclusions for pandemic related losses.
And again, it may not be what people want. But if they want covers for these kinds of events, the pricing is going to be dramatically different on a going forward basis. And the last bucket I'd say is credit and surety lines of business. This is a line which tends to re-price aggressively, people adjust exposures, and this is not the first recession that underwriters have had to work through, and it won't be the last. So that line of business I think will continue to function, but we're probably not done with losses there related to COVID.
Thomas Bohun
Thank you, Michael. Could we have the next question please?
Operator
The next question comes from the line of Thomas Fossard with HSBC. Please go ahead.
Thomas Fossard
Yes. Good afternoon. Two questions. The first one would be on the when you was coming or whenever -- you was. It's fairly abused, actually, the cyclical upturn is not led by any massive destruction capacity, and it's more driven by on the writing discipline. So, in this kind of environment, clearly the commercial strategy of the big players, like Swiss Re and Munich in the market at the start of next year is going to be very, very important. So maybe John, if you could tell us, you're thinking about playing the one-on-one annuals. I mean, are you looking for to benefit from business opportunities and capacities there, or it's going to be a mixture, or maybe there's some feel if I would say Munich and Swiss Re are going to be the one leading the market next year. And the second question would be, in relation to U.S. elections. I'm not asking you which is going to be the winner or your preference. But what could be the implications of for Swiss Re if we had a democrats coming to power or I mean, any thought on that? Thank you.
John Dacey
As I'm looking around, I think I'm the only American in the room. I might be one of the few Americans on the phone. With respect to your second question, I will retain this question on my personal vote, although I was able to vote in the state where I grew up, which is Ohio. So, I'd like to think my vote matters for this selection. What I would say, the only thing, I guess, I'd say on that one, Thomas is anything that reduces the overall level of uncertainty in the world broadly in financial markets specifically is probably going to be a good thing.
And so, what I aspire is a clarity of result next week. And if that clarity exists, I think there's a reason for us to be able to move on from the mass of getting ready for the election to what happens next. So, and beyond that, we've got a global portfolio. There is a whole lot of issues that we think about on geopolitical issues. But, sometimes it's good to be a Swiss company and things that our opportunities balance out with what some of the risks might be there.
With respect to your first question, yes, I do expect, I'm not just Munich and Swiss, but actually all the major reinsurers to be disciplined as we move forward. I think, there are two things which provide, I think, or maybe more. If I counted out and gets to four very quickly reasons for optimism for the price in the first is, and maybe the most important is the price in the primary market is improving.
And you see that on commercial insurance, which is by far the largest component of the reinsurance market. And the whether it's, major U.S. players like Chubb or the European players, they're getting rate increases for the risks they write, which allows them appropriately to be able to afford, frankly, higher prices for the reinsurance as part of that equation.
The second thing is what we've seen related to COVID losses. And again, we remain one of the more reserved COVID players, but we expect there's going to be more losses that they get booked in Q4, and maybe even into 2021. The primary companies have shown a interest in reducing other risks as they can and whether it's buying lower down on retentions and or expanding the programs that they have reinsurance the demand continues to be very strong.
The third is that the alternative capital market, which we access ourselves, I think is taking a pause, having seen a another source of loss on the P&C side coming out of COVID related property losses, which they did not anticipate. And while alternative capital doesn't shrink, necessarily, it also doesn't grow.
And yes, there's been some new capital coming into the system by some start-ups, but on a relative basis. It's pretty modest. And in many cases, that simply I'd argue replacing capital which has otherwise been burned up by the COVID losses. So, supply remains somewhat constrained, the demand is strong. And overall, the need for covering risks given the inevitable lower investment income makes us optimistic.
Thomas Bohun
Thank you, Thomas. Could we have the next question please?
Operator
The next question comes from the line of Edward Morris with JP Morgan. Please go ahead.
Edward Morris
Three questions for me. First, can you just talk about how you're feeling about the level of sort of pricing that you're seeing, relative to last trends? Obviously, on the net cap side, I think increasingly, the loss experience has been higher than expected over the last few years, and obviously there's other sources of cleanser inflation is pretty high. And similarly, a combination of low interest rates means that you really need to keep adding more and more on the underwriting side, just to understand the margin. So, I just wonder if you could talk about what you think the overall outlook is on margins, both on a pure underwriting sense and then an aggregate level.
And then secondly, I don't know if this is something you've given before, maybe you can just remind me, but I know your SST ratio has an assumption of ultimate COVID claims within it. And I was just wondering, if you could remind us what the difference is between what you've assumed as the ultimate level and what we've seen so far, going through the P&L. And then, luckily if I'm allowed to say a question, I can't remember if I'm supposed to it, myself to take. But fair question, if possible, just, could you talk a little bit around the demand outlook for some of these classes of business that have been very affected by COVID?
So, I would imagine, as you've talked about sending conditions are going to tighten a lot, I would imagine it's very difficult to buy where possible -- COVID going forward. But presumably, people can buy cover for other hypothetical pandemic events. What do you think overall? Do you think people are going to be buying more insurance than these products, or do you think, actually there's areas that it will transpire and not really insurable and just overall, how does that play out from a demand sense? Thank you.
John Dacey
All right. So, I'll take three questions because one will be a very quick answer. On the SST component of COVID, we've not disclosed the difference between our ultimate and what we booked through. Again, I think there should be some utility in the recognition that our third quarter was coherent with what we expected, when we started this calculation through the ultimate, and our processes.
With respect to pricing, I think I got to part of it on the previous question, but just to reiterate, yes, we see a need for continued price increases in part to cover what will be a decreasing contribution from asset returns. But also with respect to specific portfolio, so what we said, which remains absolutely valid at six months for our reinsurance businesses, as we saw 6% not notional increases.
Two percentage points of that were sucked up by our revisions to last models which says, we needed more rate for that, and four percentage points were covering a decrease in an expected investment income for new business. So I think, that's keeping our head above water, but it's not necessarily making this a brilliant price increase, and that's why we expected that on January 1, we'll see continued improvement in those price levels.
On the specific covers and demand outlook. I think there's, in some cases, just going to be some frustration in the market at the primary level whether it's SMEs that are just frustrated that their business interruption was not paying, due to the pandemic or other areas that potentially could involve some interesting product innovation.
But yes, we go back to the fundamental point that, you've heard, Christian talk about, which says, a true pandemic loss related to these lockdowns and the economic damage associated with it is not something the insurance industry can manage. And, our Swiss re Institute has the number out there for total economic loss of $12 trillion, more than $6 trillion already in this year. And this is not a set of losses the insurance industry can manage with a capital base we have. So, I think you're going to see some adjustments to products. But I think at the end of the day, there are certain risks which are not insurable by this industry.
Thomas Bohun
Thank you, Ed. Could we have the next question please?
Operator
The next question comes from Darius Satkauskas from KBW. Please go ahead.
Darius Satkauskas
Hi, just one question please. You said the primary rates are going up. So, the communist can afford to pay more for insurance. What are you actually seeing in terms of commissions, particularly the proportional side? Thank you.
John Dacey
So, I mean, they were -- when I say the primary industry is going up our corporate solutions, price increase, which again, is the portfolio they renewed, not what they pruned. 15%, you saw some other major players in the U.S. market talk about 15%. I think it might be a little lighter than that in some European spaces. But what we see is, across all geographies across all commercial lines, at least, some major increases were of rates for certain retail personal lines, businesses, the effects that we saw of lockdown for frequencies on motor in particular, you know, it made pricing and that those lines a little more complex.
But again, even with a motor insurance, that there's a component that that's sensitive to investment income. And we'll have to adjust accordingly. The -- when we talk about the prices for our reinsurance, it's less about specific commissions and more about just overall notional increases that we receive and/or the price adequacy again, the 6%, we saw through the July renewals was directionally correct, but not sufficient. And we believe we're going to have to see more to justify a broader expansion of our portfolio. But we're, as I've mentioned, for the reasons before, I think three questions ago, we're optimistic with those opportunities for justified price increases are there.
Thomas Bohun
Thank you, Darius. Could we have the next question please?
Operator
The next question comes from James Chuck with Citi. Please go ahead.
James Chuck
Hi, everybody. Two quick questions for me, please. John, you talked about the focus of the COVID losses it mainly on the property side of things. Just came to get your perspective on the liability lines outlook. Obviously, there's going to be lots of things that can end up in, in courts. And there's quite a long tail to this, but particularly to go into 2021. How should we think about COVID losses online such as D&O and workers compensation, for example? And then secondly, just on the investment yield. So, I think the one leg was 2.4% at nine months. How confident are you'll be able to sustain that level? Can you give us any guidance as you reinvest excess cash? What's the kind of headwind in terms of basis points on that on that investment? Thank you.
John Dacey
With respect to your first question, I guess I won't be surprised to find some disputes and potential litigation actually on the property book as well in terms of a disagreement over what an occurrence is disagreement of how accumulation works for the treaties under which losses might be claims. But we'll see how that plays itself out. On the specific set of liability lines, I think the one place which has been reasonably well-behaved at least now is the U.S. workers' comp. It's not an area that we've got a big exposure, but there has been some major decreases in premiums that have been captured by the insurance industry due to the fact that the workers had been for -- positions closed to the degree of the economic recovery is recalling some of that. It seems to be functioning okay.
I think the D&O side, yes, there's likely to be some losses coming through, again, not a big exposure necessarily for us. But even that's probably been mitigated by the relative rebound of share prices in particular. And some of the losses will come from people that bailed out in April and didn't stick around for some of the rebounds. But we'll see how that plays out. We ourselves have a material reserves set up for liability lines related to COVID on the CorSo side. In particular, we've not seen much at all in terms of reporting losses. And so, that remains largely in IBNR position. We'll see how it sorts itself out.
With respect to the investment yield, yes, there's a 2.5%, 4% is down. Our current re-investment yield is looking to be about 1.4% of the duration of the portfolio approaching six years means that it's going to take time to migrate down to that number in between now. And then things can change, obviously, but at the moment, I think consistent with the earlier question. We're not going risk on to try to maintain it. We're going to be cautious at least until some of the uncertainty that we're facing, and the economies and the political situation in multiple countries, not just the U.S. resolves itself a bit. And so, we'll see how it works.
Thomas Bohun
Well, thank you.
