Up next we have Arch Capital. We've got Marc Grandisson, who is CEO, just rounding the corner of one year. François Morin is CFO up here as well. For me, one of the nice things about being around for a long time, and I think I'm older than I look, I hope-
Not that old
because I'm pretty old, is that you get to see the evolution of a company. I actually worked on the Arch IPO back in 1995, so I've been witnessing this company's really amazing journey. Marc has been with the company since its founding.
Has been a critical part of the management team, and he helped build a company with a unique, and I'd say, successful culture. That culture is an asset. Doesn't show up in the balance sheet, but it shows up in the results over time. It is great to have you here, Marc-
Yeah.
François.
Thanks for giving us, I think, for the last couple of years.
I always say nice things about you.
Great.
As I said in the intro, you've been CEO for about a year.
Yeah.
Give us some feedback, what you learned during that first year, and maybe just to add to that question. You've watched a CEO for many years.
behave, and you probably have some observations on that and how he did it.
what the challenges he faced. How is it in the seat versus watching that person?
The lights are on you the whole time. One thing that I learned throughout the year is that you better know what your principles are and what your true value is at the core, so that you know how to react. You also have to stay calm through all this. Last year was very interesting in many ways, one thing I learned, and that really surprised me over the last year, Jay, is the amount of work and education we still have to do about our MI business. It's extremely difficult to appreciate from Even though I was still part and parcel and side by side with Dinos and the team to explain it's amazing how much time it's taking for all of us collectively to really get our arms around that business.
Being a P&C, a lot of our investors are P&C-focused, it's, as you know, a big portion of what we do and big portion of our earnings and the growth that we've had over the last two, three years. It's been difficult. It's been an interesting year in MI, at least from a headline perspective. The one thing that's really difficult sometimes as a CEO is to get the noise away from your brain, because there's a lot of noise about the MI business, be it price, competition, credit. A lot of things are said out in the marketplace. What we have to reconcile is a lot of times what we hear in the marketplace and what we see in the data, I know we are a data-driven company. There's a huge disconnect, and there's been a big disconnect over the last several months.
I think it's getting a bit better of late, this has been difficult, to stay away from the noise and really stay sturdy and look at the numbers and factually base your decision through the year. It saved us over the last 18 years, and I think it's going to carry on in the future. That's really something I learned from Dinos, we talk about that all the time. In the short run, you heard this from Warren in the past. In the short run, it's a voting machine. In the long run, it's a weighing machine. Stay the course, I guess, is what I've learned. Very close to heart and very right front and center for the last 12 months.
Has anything been harder than you would have expected? You've been part of the management team for a long time.
Yeah.
You've been involved with every key decision of the company.
Has anything been tougher than you thought it might be?
No, not from the inside, not from running of the company. A couple of folks left, some retired, some went into different positions, different opportunities, and we're happy for them. A very deep bench in the company. People that have been around. I've known this guy for 33 years, went to college together. The guys who run insurance, reinsurance, and even the mortgage, I've known for 20, 25 years. Very stable group. The insurance group transformation, a couple of changes of late. All guys that have been with us for a long time. That's been good. It's actually refresh sometimes the idea and the outlook on life has been a positive, I feel, more than negative over time. You see this happen. No, I haven't seen really something much more difficult.
I think that the market conditions have been more harder than I would have expected them to be. I would have expected them to be, I'm meaning softer than I would have expected. That's just what I meant.
More challenging.
More challenging, yeah. I would have expected some of the events of 2017 and currently as we see it in the events of 2018, allowing the industry to make a statement about what needs to happen in terms of pricing and return on capital allocated to the broad capital allocated to the cat space. We would have expected a bit more of a hardening. It just didn't happen. We still haven't seen as much as we would have anticipated, hoped for as January 1st. That's been a bit more difficult, but that's external created by-
Right
the supply and demand capacity.
Let's start on that subject. Looking forward, you think the market's going to wake up and firm up this year?
I don't know is the short answer. It's a very difficult thing to predict. I think we all collectively probably have called for like 25 hard markets and only three of them occurred. I think what we hope for and what really happens are two things. All we need to look at is how much capital is allocated right now, how much has gone out of the space. The demand is pretty inelastic. It's pretty stable through time. People buy and need the same amount of capital on the cat side. Unless times like 2005, where the modeling gets a bit more punitive and then you need to deploy more capital, and you have a recognition of more capital needed to support the business. Right now that is not happening.
There are different people that are desiring or are willing to accept different rates of returns out there. Some of them are lower than others, and a lot of capital that we see out there is willing to live with a lower rate of return, and it's not going away. Therefore, the overall blend of the return doesn't change a whole lot. It seems to indicate, unfortunately, that we're not going to have similar change in the pricing as we would still hope. Having said all this, maybe people change their mind. Maybe there's some things that get reassessed between now and mid-year, but we'll see how that goes.
I think it's for Arch. You're such a different business than other companies that it's a less relevant question. You'd like things to get better, but you're not clearly dependent on it.
No, we're not. You're quite right. We'd love to do way more of this. We used to do four or five times what we're doing right now in terms of premium written and exposure certainly, a fair amount, much more, five times almost what we're doing right now. You're right, it's not as crucial to us, even though it's still something that we can offer to our clients, and we'd like to offer more at the right price.
Let's talk about mortgage insurance. One of the issues that came up in 2018 was an increased concern over credit, all kinds of credit. Here you are with a mortgage insurance company, arguably exposed to credit deterioration. As the year went on, did you begin to adjust your underwriting standards in that business?
Not really. I think that the risk-based pricing, of course, you always update your risk-based pricing. Our RateStar is a dynamic process of reassessing, evaluating where the price should be and what kind of return capital you need for certain segments of the portfolio. That's an ongoing one. It's not specifically related to what's happening 2017, 2018. I think in 2018 to even into 2019. The credit riskiness, it has a little bit expanded over the last four or five years, and we won't argue with you, but it is still such at a healthy level that you cannot compare where we are to pretty much anything that we see historically all the way back to the mid-'90s. It's still better. There was a big shift in the fundamental of the business after 2008.
Not only did the credit worthiness, the credit box we talk about got tighter at the bank level, and then everybody's really being careful the way they extend credit for housing. The level of scrutiny that the MI industry has to go through to provide that business, to provide the mortgage insurance policy, and also what has changed is the amount of capital they need to set aside if they do provide an MI policy, is actually a little bit of a revolution, even for us in the P&C industry. Starting in 2008, we started working on this, and as of 2013, 2014, the PMIERs came into play. This is a capital regime that the GSE and-- I'll circle back to your credit in a second, but the GSEs have created a countercyclical model unlike anything else we have.
If you look at the P&C capital model, you get charged for premium you write, you get charged for the reserves you book, you get charged for the investment, of course, that you have at your risk and your portfolio, and you get charged for the cap that you are taking on the balance sheet. Three of those four, it's made up for you to underreport those numbers. It sort of incentivizes the companies to underreport because if you over-report the premium or the reserving, you get penalized. If you switch that, so it's not a very, I believe, aligned with the way the cycle would work, right?
Yeah.
The reserve ought to go up. You're not going to get it up because you're going to need more capital to set aside to survive that test for the capital agency, for the rating agencies, or for the regulators. If you look at the PMIERs, if you write something right now, if you go from a 760 FICO to a 740 FICO, you need to hold 30% more capital. In terms of you willing to extend credit, you have about a 30% increase in capital you have to put aside. If you have no documentation, that 1.3 gets multiplied by three. You need to have four times the amount of capital as if you wrote a 760 with full documentation. You have embedded in this capital regime a countercyclical mechanism that you can't hide from it.
There was nothing like that before.
Before you would do a house, a $100,000 home. You're a FICO of 795, just looking at you.
792.
792, right? Looking at it every day, making sure you're good. Trump, on the other hand, is a 520, right? Before you had a $100,000 home, a $100,000 loan. 25 to one capital for you, 25 to one, a 4,000 charge. In this day and age, your 4,000, it's probably 25,000. Whole different segment, all different argument. It was very similarly priced, right? There wasn't that much of a difference on the rate card, not as much as you would've hoped for.
Right.
Risk-based pricing puts everything in that perspective in the right format. Yes, we can have competition. You can have credit getting slightly worse. The question that we all have collectively as an MI industry have to ask ourselves is, are we willing to take on that increased credit risk? It's going to mean additional capital that we have to put on our balance sheet. That's what you saw at the early 2018. End of 2017, early 2018, as a result of the Fannie 50% DTI, more above 45 if it gets fed to the market. Some pressure to get more higher than 95 LTVs in the marketplace. Some people get a lot more than they ask for.
Still price, you could argue that it's still a very interesting loss ratio, that business, a capital charge that those guys were facing said, "Wait a second. We got to take a break. Got to step back and say, let's reassess what's happening." In terms of creditworthiness, a couple of facts for you. Jay, I know you are a student of history.
If you look back at the early 1990s, early 2000s, I said some things in my comments yesterday-
on the earnings call. If you look at the LTV, DTI, and FICO, right? If you look at where we are currently versus where we were in 1999 to 2000, where the loss ratio was about 20% for the MI industry, this is pre-regime, pre-change of regime, right? The LTV was about 95 on average for the LTV above 80. It was about the same level. Sorry, 1992, 1993, the same way we have right now.
The DTI was about 36, 35, 36, the same that we have right now. FICO is a big, big, big difference. FICO back then was about 705, 710 on average. Now it's 743, right? FICO, and that's in this market that you call is credit softening. It's still an extremely high level of FICO. The additional thing that I would tell you that is also not present in the marketplace as much as two more things I can go all day on this one, is our risk layering. Risk layering is doing an LTV higher than 95, a FICO less than 700, let's say, and a DTI above 40. Back in 2000, it was about 2% of the production. It's 0.2% of production as we speak right now.
It's nowhere near the level. You should ask me in 2007, 2008, where did it go? That risk layering went to 3.5%-4%. The additional thing I will tell you on creditworthiness, because I'm not done yet, is pre-2008, 2007 and 2008, the product was different. The mix of products were different. You have none of the product that we call the Alt-A to subprime, so the liar doc and the no-doc. None of this happens. Why? Because if you have a no-doc, it's $3,000 amount of capital. That's not there, nowhere near there. Even though there has been some widening of credit to some extent, which I said yesterday was still 100% better than we were back in 1999 to 2001 sort of period, where we ran a 20% loss ratio, we're still far better than this.
You cannot compare this current credit environment with anything historically, because the tight in 2013, this is where the tighter market was the tightest, is nothing you can compare. These are all numbers I gave you. In addition to all this, is the appraisal is so much better than it used to be before. The LTV I quoted to you in 2007 was about 93. Believe it or not, same LTV that people were producing. The appraisals, all you guys know here, the appraisal industry was not doing its job really properly, right? They were really writing something at 110-115 LTV, but declaring it as being 95, right? Documentation, I just mentioned about it. Even the DTI now, what gets accounted for in income is a lot less than before.
Somebody is an Uber driver and works full-time job and has Uber on the side five days, six days a month, don't get to account that income necessarily against its income for the DTI ratio. We're talking about a sea change in terms of creditworthiness. Even if we have a small change in credit over the last two, three years, it's just so far above anything we look at. We have this 1999 to 2001 sort of history. We have a better environment than this one, this is where it ran.
Right
pre-regime. Matching the headline that we hear about, that you just mentioned, that credit's getting worse, and the data is just a huge disconnect.
Well, I think the headline really wasn't credit's getting worse, credit may get worse. The fear was credit would be getting worse.
It gets worse. We react accordingly.
Yeah.
We price for it, we don't get the business.
Yeah.
We still have this beautiful business we wrote for the last six, seven years that's still.
Right
Giving us premium.
Generating income.
Building equity further away from the color gango.
Let's do a scenario. Let's say the U.S. goes into a recession. Not a repeat of 2008, 2009, but a normal run-of-the-mill recession. I know they're all different. Given how you're positioned, given everything you just said, what happens to your MI business? That would be instructive.
Well, if the scenario is one where it's, again, not national, it's a bit more regional, right? A, if you look at the performance, there's a lot of premium that's still going to come through and hit our books. Let's say the scenario, it's relatively adverse, but it's not outsized. Maybe the loss ratio gets moved from the low teens to maybe 30%, 40%, 50%. That would be, again, fairly sizable event. What we've done, we've bought a lot of reinsurance protection, and we've disclosed that, made that very public. Our Bellemeade transactions is something that our friends at UGC, before the acquisition, had placed two of these types of deals before the acquisition. Since then, we've made that more systematic. Every six months, we go to the market, and we've actually done a few additional one-off transactions.
We go to the market twice a year and really place a, call it a mezzanine layer of risk.
Which typically protects us, in loss ratio terms, roughly 25% up to 75%. Right there, we're capping off a fair amount of the downside that we have. No question that we could go over the top, but that type of scenario would be truly, it'd have to be really a broad-based national kind of event that would be similar to a repeat of 2008. Anything that's less, a bit smaller or not as well spread out would not be as damaging. We think we get, from our balance sheet point of view protection, I think we're in a good position. Again, what does it mean to the business? The question then becomes, well, what happens to the competition? I think the competition, if it's not too big, I think most of our peers are doing similar kinds of things.
They're no longer just accumulating risk, which is what they were doing up till 2008. Back then, the model was very much buy and hold, and cross your fingers, and everything will be good. I think everybody's come to realize that you need downside protection. Presumably, the industry would be healthier, and there wouldn't necessarily be any insolvencies or any of our competitors or peers that would have to shut their doors. The question then becomes, well, do we reassess the risk, and does pricing adjust, yes or no? Maybe it does, maybe it doesn't. That's a bit unclear.
I think, again, like Marc said, I think that the business, the model itself is just systematically, fundamentally different than what it was back then in terms of risk management, in terms of balance sheet protection, in terms of having a more, again, the black box, the risk-based pricing that we have that others are making available now is also making a difference because there's more segmentation in how the business is priced and the returns we're generating on that business.
If you look at our risk-based pricing, Jay, and I think it's going to be the same for our colleagues and our friendly competitors, is that as you develop the model, you also get some regional differentiation in terms of pricing as well, right? It's the one part. You can also adjust for, you see some pressure points in one area of the market, you can actually adjust your pricing to reflect. You can take pro-action as opposed to just receiving it and do the best you can. It's actually a very different reaction. It's more pro-action ability than before.
Given these changes, you look at the public MI companies, not you.
Yep.
The multiples they're getting aren't particularly good. They trade pretty cheaply. Do you have a sense, it's not a fair question for you guys, it's not your job, but any perspective on why you think that is?
Maybe, Michel.
I think it's still the fear.
Yeah.
You just said it. People are worried about price. People are worried about competition. People are worried, and I can understand that. People are worried about these guys don't know what they're doing. It's the same old you know what. They're going to do the same thing they did before. That resonates with me. It's still very fresh in everybody's mind, and some have a lot of wounds still to show for the crisis of 2007 and 2008. It's a sentiment, I believe. A lot of it is sentiment issue.
Yeah.
I can see why people feel that way, I would also tell you, Jay, that's also why the industry can generate those kinds of returns. There's still a lot of fear about what's happening, I think the more you scratch the surface about what I told, and we talk about the model, I think that would alleviate the fears over time. People always tell us that we need some kind of crisis to retest the system again, right?
Yeah.
That hopefully doesn't happen. That's what we're hoping for collectively, not only for us as MI, but as citizens of this world. There's a lot of fear still.
Yeah.
It's hard to go back, go in front of your portfolio manager and say, "I want to buy the MI. I think it's undervalued." People say, "Well, isn't that the same industry that blew itself up 10 years ago?
Yeah. Go away." I can see that. I can see that.
Any questions from the audience for Marc or François?
No.
I wanted to oh, we got one.
We got one.
Just wait for the mic, Paul.
When the market got hit last month, I know you bought a couple of small companies and used cash for that, but I'm surprised you only bought $100 million worth of stock given all that you're saying about MI and what the cash flow is that you're generating. If you look at what your stock traded at that time, and look at the cash on cash returns on cash earnings.
where else are you going to find something as attractive and.
Well, that's a very good question. Part of it, we were in the closed window, once we set the grid in the 10b5-1 plan, we can't do anything about it, we just got to let the formula get to work. We knew these acquisitions were coming up. We always all along had flexibility. We paid down debt as well as revolving credit facility. We did a lot of things in the fourth quarter that touched on debt, touched on stock repurchases, touched on acquisitions. There's a lot of things we did. There's rating agency considerations as well. We just got to navigate through that. Could we have bought more? It's easy to say that now, obviously, because the stock's gone back up in.
Short time.
I'd say all I can say is really it's something we look at every day, every month, every quarter, and we're going to keep doing what we think is right.
Our insurance in force is growing. It's growing this quarter.
Yeah.
It's still growing, and it's still very good returns. That's also competing. There's competition for capital allocation in our company all the time, and it's not like all one or the other, right? We're also an insurance company, right? We're trying to build also for the future and invest in some things that may not necessarily be hitting the earnings calls, but we also are trying to establish ourselves, positions, investment in things for the future as well. Everything is competing. We try to do the best we can, I guess, is the answer.
On the other side of the coin, the MI business, as you said, the way it's cash flowing, if you decide to be more conservative in writing new business, it's a good problem to have.
Yeah.
You're going to have a nice melting ice cube.
That's right.
You need to do something with that ice cube.
Yeah. Also, in general terms with the MI market, I want people to understand that, even though the mortgage origination is planned, it's projected to go down next year, right? In terms of our target market, which is really the purchase market, it's actually going up for the next couple of years. There's going to be more increased demand, demographics as well. Well, that's the other thing we'll go into in a second, in general terms, the purchase market is growing for the next three years. We'll have to put that into consideration when we look at capital deployment.
Any other questions? Just to move off of mortgage insurance, I'm glad we spent that time on that. I wanted to ask about Arch Insurance. Obviously a tough environment. You guys have always found pockets of opportunity to grow. What areas of the business excite you, and where could premium growth come from this year?
What you saw in the fourth quarter is a good example for what we're seeing. There are three areas where we're growing, and for different reasons, some of them, right? Travel is one that we're growing, and that's sort of the ongoing buildup of that capability within Arch and a couple more relationships that we struck that are paying off, and we're growing that book of business. Some rate, but mostly growth of relationships. Program business is also something we're growing, and that's really a result of rate change, rate increase, and then change in some products that we're offering them. By and large, it's a stable program book that's growing in premium, so we're growing in there as well. The third one that we've grown is the property and the marine, and that's really property. That's pretty rate driven. That's really cap influenced and impacted.
That's creating for us opportunities on the insurance side, right? We also feel a similar kind of rate increase in pockets, as I said yesterday. It's not across the market, but in pockets for the reinsurance group as well.
Yeah.
That's where you see us grow. We have a couple growth area here and there that don't show up because D&O in the U.K., for instance, might have some issues last year. A couple of things, a couple of opportunities to grow in that area as well. By and large, what we like, you see the growth in premium or decrease in premium, that tells you where we think the market is.
Yeah.
To add to that quickly on the acquisitions we announced late in the year, I mean, that's a bit of a new territory for us. We're trying to be a bit more retail both in the U.S. and the U.K. That's something that we're going to make a push into. It's not going to be, again, transformative, at least not initially. We think that's a way for us to get access to a bit more distribution and be closer to the customer. Not only use the products that they already sell, but offer some of our existing products through that distribution channel.
Yeah. Last question from me. There's changes at the top over the past year. Do you have the team that you want right now? Are there spots that you think you should be adding to from a people standpoint?
I mean, we're always looking for great people. We have great people. If we have other great people, we'd like to take them on. It's an advertisement for the whole world. If you were looking for a new home, we'd love to talk to you. All kidding aside, I think we're very pleased where we are. I think most of the team is intact by and large, and I think if you look at the people who run the units, it's a very Arch-like way of thinking and culture throughout the executive team and throughout, down the organization. Yeah, no, nothing really to speak of, really.
The way asset management is going, you might get 10 resumes.
We'll look at it.
Why don't we cut it off here? Marc, François, thank you very much.
Thank you, Jay.
Thank you.
Thank you, guys. Thank you.