I'm going to mute you off. Okay, we are going to get started in the hopes of staying as much on schedule as possible. With me for this panel is John C. Roche , the CEO of The Hanover Insurance Group after a distinguished and impressive career at Aetna. I'm going to start with a broad question, that is with regard to the Hanover 2021 strategy that was laid out earlier this year. If you could hit the highlights of how that's going so far, I know it's early-ish. Maybe we're, I don't know, 10% of the way in on that period, but you've had some notable successes, and I was hoping you could talk about that.
Great. Great to be here. Thanks for having me. Let me reiterate the sort of core elements of the strategy. At our investor day on February 23rd, we outlined a strategy that leveraged the best of what this company was good at, but doing things slightly differently. At the core of the strategy was to drive hard at the independent agency channel here in the U.S. by growing market share from 7%-9% with 2,100 deep relationships. I can take you through the six and seven things we are going to do to actually drive that growth. Second part of the strategy was to better leverage an asset that is really not well-known, that is that we're in the specialty business to the tune of $900 million here in the U.S. It's a fast-growing book.
It's a profitable book, yet it's not fully penetrated with the independent agency channel. So if we can leverage it with the IA channel, if we can develop a wholesale and E&S capability, a new distribution system for that product line, and by working with John Fowle at Chaucer, figure out how to synergize both U.S. and global specialty, we think that's another winning element of the strategy. Third, since we are relying on distribution partners, both London market brokers and the independent agent channel in the U.S., we thought that helping our distribution partners innovate in an ever-changing world that puts them at some risk of marginalizing components of their business that are in the channel today that could leave or never show up.
That means getting at micro commercial in a more direct way, pulling the millennials into the channel that might otherwise seek products in a differentiated way, et cetera. Drive hard at the IA channel, leverage our specialty capabilities, and help our distribution partners innovate. All of which, in the context of maintaining a combined ratio at a profile we enjoy today. 7%-9% premium growth, a combined ratio of 95% or less, book value growth of 7% CAGR to get to $100 of book value per share in 2021 with a double-digit ROE. Sounds like a heavy lift, but that's the strategy as quickly as I could articulate it. I think the first two data points in the quarters were actually suggestive of the fact that so far so good. You never like to declare victory on a five-year strategy with two quarters of results.
If you look at our small commercial Personal Lines and specialty results over the past two quarters, the fact that we were able to grow those books of business, hold serve on our combined ratio, increase our persistency, grow market share with our agency force, is testimony to the fact that at least in the early stages, the strategy appears to have traction, is probably the best way to articulate it. Still early, I think the strategy has traction.
Thanks. In general, I'm going to start off-- wow, this is loud. Start off with a few introductory questions. When there are questions from the audience, again, obviously the conference is for you, so just let me know and I'll be happy to turn over questions to anyone on the floor. Let me start digging in a little bit on the net written premium growth side. Again, the goal is 7%-9% CAGR taking us to $7 billion in 2021. There's a fairly lengthy list, even in those two quarters, of accomplishments, right? There's the Hanover Fusion life sciences product, entering Pennsylvania Personal Lines, forming Chaucer Dublin, the McGowanPRO accountants program, the acquisition of SLE Holdings. Not all of these impacted year-to-date premium production, which is about 4%.
We put all of those pieces in, again, I recognize it's an unfair question because it's early. Are you ahead of schedule? On target? How do you look at that, again, the premium growth component of the 2021 strategy?
At this early stage, I would say we're on target and feeling good about the traction, you raise some interesting points. We're never going to grow the way we say we're going to grow without having the flow business. That is the daily transactional business where you have your product on a shelf in an agency. They understand the breadth of the product line really well. You have stability of capital in that market and the stability of pricing. It's all about the flow business. Step 1. Step 2, market consolidations. Agents are willing to convert major portions of their book of business to a carrier who helps them win. Whether it's deconsolidating the top, where a carrier's gotten too much market share and has too much financial control, or consolidating the bottom, where the bottom 20% of their premium is fragmented across hundreds of markets.
We are the best at helping agents consolidate major portions of the book and moving it from a set of carriers to Hanover. That's sort of growth strategy number two in terms of its execution. The ones you mentioned are absolutely in scope, and that is what I call pseudo M&A. You can always go out and try to buy a company for 1.8x book value and try to put up with the dilution, but what we like is we like buying MGAs and converting their premium to Hanover or Chaucer paper. We like renewal rights deals like we did the OneBeacon deal seven years ago and the deal we did with Encompass in Massachusetts just a few months ago.
Whether it's a renewal rights deal, whether it's a conversion of an MGA, or whether it's these mega consolidations we're doing with agents, the items you mentioned are a major catalyst for growth. It will add to both the transactional flow, the consolidations, and what I call pseudo M&A, and we've got a great business development team that knows exactly how to execute those types of things. We'll enter new states when we have to. I've told the team, until somebody convinces me that the shelf is filled with every single agent relationship we have, same store. New store, same geography. Are there more appointments we need in geographies where we're already in? After somebody convinces me we're filled up and leveraging that existing fixed cost structure, then we'll start to entertain new states.
We don't have optimal market share in the areas where we're dense, and we certainly have a few more appointments we can make in areas where we're less dense. Let's fill those buckets up first before we go off trying to launch new states.
I want to talk about that a little more. Can you talk about the pipeline for either additional agents or additional shelf space in the current agents? In other words, how you score those trends, what the opportunity is and where you are.
Our analysis is rather empirical and bottoms up. We did market share analysis agent by agent. It's rather intuitively obvious that the bigger the agency, the lower market share you have, and with some of the smaller guys, your market share could actually be 20%-25%. On average, it's only 7%. When you look at the top three relationships they have in a market, sometimes those are 20% and 25% relationships adding up to 65%-75% of their book. They're not going to let those get any bigger. At the bottom, it's hugely fragmented, where they have hundreds of markets that have been orphaned over the years, where they couldn't understand the coverages, they didn't understand the pricing dynamics.
We look at single digit market share and saying, I only need to go from 7-9 to get the lion's share of that billion and a half dollars of premium. We're not sitting around wishing it to happen. It's by being slicker on the front end, opening up our risk appetite to conform to the market. It's by doing these mega consolidations and by driving premium growth in the specialty lines where right now I would say we are woefully under-penetrated where we should be compared to the size of the standard lines business we have with our customers. There's four or five very discrete initiatives that we attack daily to execute on to make sure that we can hit that 7%-9% growth. As I said, it's only going from 7%-9% market share. It's not going from 7-20.
It's not going double digit. Going from 7-9 with 2,100 very special and intimate relationships we have with our distribution partners.
Okay. When you talk about the opportunity in specialty lines, is that to date the under-penetration a function of products at Hanover, of brand awareness?
That's an excellent question, by the way. Let me clarify. That under-penetration is just analyzing the level of penetration we ought to have with our existing product line. The way you do that is one of the aspects of the relationship we have with our agents is they generally give us access to their data. We have a tool called Agency Insights, which is a proprietary set of algorithms and analysis tools that helps the agent actually understand the composition of their book of business and how it can be repositioned to help their customers, help themselves, and help Hanover. By running that analysis across our entire U.S. distribution plan, it's the Amazon analysis. Customers like these have. You know that if there's this 20-person accounting firm that has a package, they also have a professional lines policy somewhere.
They also have this, they also have that. We know that our specialty product line is three to four times under-penetrated in relation to the amount of standard lines package business we have with our customers in our existing plan. That $400 million of premium growth over five years is with no introduction of new product. Current products, current plant, current customers.
Just addressing the under-penetration that developed then.
It's really interesting because I wish I had sort of a more intellectually interesting answer for you. It's muscle memory. You go out to the field, and I'll tell you, our field guys, our field organization is one of the best in the industry. These guys know the product lines really well. They better represent the standard lines packages and the personal lines package because we have both, and less skilled in construction, surety, less skilled in programs, less skilled in professional and management liability. All we do is build up the girth in the field to make sure our product lines are better represented. By the way, it's not with 2,100 agents, it's with a subset of 2,100.
Right.
Not all agencies have built the specialty capabilities to go after surety business, to go after inland marine business, et cetera. We know exactly where these agents are. We know exactly which ones had chose to specialize, and we know exactly what we need to do to better represent our product line in the field. The agencies routinely tell us that I have this product with this competitor because five years ago when I needed to place it, with Markel, Beazley, whoever it was, they were there and you weren't. If you have the product line and your pricing is rational, I'd actually rather place it with you to round out the entire Hanover relationship.
Okay. Let me ask, you sort of touched on this, but I want to make it explicit because I think it's an important part. When you had the presentation, when you had the investor day, you compared the 7%-9% CAGR with 7% between 2011 and 2016. One notable difference is that for a lot of that period, pricing was more favorable. You had the benefit of rate increases from, let's say, 2011-2014. Maybe once we get to 2019-2021, we'll have favorable pricing as well, but as I understand it's not built into the expectations. How does that factor in?
I think you have to parse the portfolio. I think when you start painting the portfolio with a broad brush, you can actually make some intellectual mistakes. Fair to say, pricing in the marketplace, generally speaking, is pressured. It's hard to put rate out there that's keeping pace with loss cost inflation. No question. Let's parse the portfolio. Right now, we're putting 4-5 points of rate into our personal lines portfolio. That's 100 basis points ahead of cost trend. We're starting from a very good pricing point, so it's not putting pressure on us. Everybody's trying to catch up. We've routinely put 4-5 points in the market. The quality of the book of business is improving to where the retention is moving from 85% closer to 90%.
Every day we write new account business and new platinum business, that retention ratio is inching up. You start putting 4-5 points of rate on top of loss trend 100 basis points lower into a book that's being retained at 85%-90%, you're going to grow and you're going to grow profitably. We're feeling really good about that. While in Commercial Lines, there's no question that the premium you're able to get is not keeping pace with loss trend. That is more true in middle market and less true in small commercial. Small commercial is not a terribly price-sensitive business. Small business owners like to place the business, be absolutely reassured that when something bad happens, you're going to be there, but they don't want to touch it for three years.
As long as you have stability of pricing, and pay the claims when they actually are due, you can have a very stable book of business. The negative spread between trend and yield is less true in small commercial than it is in middle market. Specialty is specialty. Highly specialized coverages, sophisticated risks, high value add, less price sensitive. That brings us to middle market, which is the battleground. There's a lot of capital in the middle market chasing very few risks, and pricing conversations start with how much reduction can I get. It's a battleground, but the agents need to play there. You can't be relevant with an independent agent unless you play in the middle market.
If you can hold your combined ratios in the high 90s and get all the other stuff that comes along with it, like personal, small commercial, you can do very well. Your point's well taken. It's clearly more pressured today than it was a few years ago. If you look at where we are putting our capital to grow, the price sensitivity is less onerous in those areas than it is in others.
That's very helpful and very thorough. One question I have, I guess, is I think you're absolutely right in the theory of specialty insurance requiring specialty insight that customers will pay for. It's not always the case that your competitors are aware of that, and specialty seems to be the first port for companies that are deploying capital away from property catastrophe. More compact distribution force.
Sure
The opportunity is there. I'm not disputing at all with your assertion of what should happen. I'm a little surprised that is actually what you're seeing happen.
Yeah. I forget what the first quarter numbers were. I actually don't remember, we grew, if you pro forma out that reinstatement premium we had, we grew specialty at 6% in the second quarter. That's after we had to reposition the surety book and the program book based on the underwriting snafus that took place two or three years ago. This is real. I mean, inland marine business, contractor builders risk, contractors equipment, PL, ML. It's important to the client base. It's specialized, but it's not so complex that we can't keep driving it through the independent agent channel. It's specialized. I call it specialty light.
It's just less sensitive to price. It's not insensitive to it's just less sensitive to it. A certain subset of our agents focus on it as an area of growth for them. Keep it out of wholesale and keep it in retail.
As soon as it goes into wholesale, they lose five points.
Right.
They're finding ways to try to keep it in retail. We may develop a wholesale E&S capability. Bryan Salvatore is being charged to do that. If we can't, all we need to do is keep penetrating that product line through the IA channel. I think we can grow it at mid to high single digit rates, and I think we can hold our combined ratios, particularly in inland marine at 90 or below where they've been for the last six or eight quarters. I really do.
Okay. Again, if there are questions from the audience, I'm going to regularly look up, to give you an opportunity to have a break from me. In the absence of that, I'll keep going. I'm going to ask the same sort of parsing question on what we call the core or the accident year CAT loss ratio. Because one of the goals is to keep that flat for five years.
Right.
Far, I've got my notes so I don't get it wrong, we've seen some improvement in Personal, a little deterioration in Commercial, and Chaucer's been stellar. If we go through the individual components there, you did touch on the Personal Lines side. You're getting rate above trend, that's kind of easy. I'll put that in quotes. What about the sustainability of either Commercial or Chaucer's results so far?
I think in the Commercial Lines side, again, putting the two portfolios that had to be repositioned aside, the real question, not just for us, but I think for the marketplace was is the marketplace observing a casualty trend that is moving outside of pricing. In down economies, you see more attorney involvement, you see more soft tissue claims, you see more medicals going in early. You see all that. The question is, how soon do you see it? Our antenna is up on casualty severity. Sometimes, you make the most out of a tough situation. I knew we had to take the reserve charge we did at the end of the year.
Which caused you to actually go in and unpack the underlying loss trends in your casualty business and say, "What is the real trend here? If it wasn't recognized in the past, I better understand what it is now." Here's the chain of events. You book your reserves on 2015 and prior to the level of trend you believe to be true with all of the factors I just mentioned, increased attorney involvement, more soft tissue injuries, et cetera. You then immediately look at your current accident year and say, "Does that 2015 and prior trend inform me better on 2016?" What did we do in 2016? Look at the CMP loss ratio for the fourth quarter. We popped it because we thought the trend we were booking into the current accident year loss ratio did not reflect the prior trends.
The next thing you do after you do that is you then go in and look at the pricing models you have out there for 2017. I'm telling you make the most out of a crisis, you book the reserves, you completely take all that information through your entire book of business. I would tell you that we just had a completely fresh look at the end of last year at making sure the underlying average severity assumptions on casualty were baked into our results. We think we have a really good starting point, and that the pricing adequately reflects all the trends that we recently observed as a result of our reserve review.
Thanks. I have to say, this is a comment, not even a question, that the investor response to your reserve charge was something I'd never seen before. It was, I think, a phenomenal vote of confidence in terms of how a new management team, relatively new CEO, brand new CFO, how they were going to run the company. It's absolutely the right call, and I think the market recognized that. Talk a little bit about workers' compensation. It's a big line of business. A lot of uncertainty, I guess, with regard to medical or the healthcare system legislation and so on, but your results have been absolutely phenomenal. I was hoping you could sort of talk about what's driving that individual line of business.
Yeah. Without being too glib about it, I would say the reason the performance is where it is we've been very selective as to where we play. It's only a $300 million book of business. You run out of opportunities in technology and in human services where the coverage is compulsory, but there's no claims. It's just not a claim-intensive segment of the industry. We play in the white collar, not-for-profit technology space. Product is really important for us. The key question is, if we moved up market and had a broader portfolio of workers' compensation, what would the results look like? Our agents are encouraging us to be slightly more expansive in our risk appetite. That doesn't mean we're going to start doing longshoremen, and haulers.
When they're placing a package for a mid-size manufacturing company, they don't want to bust up the account. Sure, you want the package. That's the easy part. Take the commercial auto, take the workers' comp, and give me an account relationship that I don't have to bust up the account and have three carriers supporting my client. When we said at Investor Day we were going to conform our risk appetite to the market, generally referring to commercial auto to some extent, but workers' compensation, move a little up market, be a little more expansive in terms of what we'll take, all with the goal of having an account relationship that still produces 95 points or better. If the comp produces 102, do I really care as long as the account produces 93? That's sort of the way we look at it.
You're not going to see us go up market into the loss-sensitive business, where we have to control pharmacy and DME and have thousands of back to work case managers, nurse case managers. You're not going to see us do that. Moving slightly up market to give our agents the ability to place the entire account, I think is really important. By the way, we're actually, the combined ratio would suggest, really good at least with what we do today. To be honest with you, it's really a function of the places where we play rather than we're outperforming. We're playing in a place that gives us the ability to produce those results.
Should I view that in the same light as your earlier comments on the middle market? Is this middle market workers' compensation we're talking about, or are those two similar strategies of winning broader accounts, broader agency profiles?
It's both. I think about 55% of our workers' compensation book is small commercial, the rest is middle market. In middle market, it's at the low end of middle market. These are not $500,000 accounts and $1 million accounts. These are $50,000, $60,000 accounts. I think obviously, if you don't chase exposed classes, you're not going to get any premiums. You have to chase exposed classes, do it in a responsible way, all with the view of rounding out the relationship for our agents. When I say move up market, move slightly up market and be more expansive in the asset classes or the industry classes you're willing to write for small and moving slightly up market in middle market, staying away from anything that gets close to the loss-sensitive type business.
That's helpful. Again, obviously, if there are questions, please let me know.
Sure.
Yes, Steve?
I know it's early days, but what can you share about your thoughts surrounding Hurricane Harvey and the exposures that the company has? Sure. The comments we'll make today on Hurricane Harvey will generally be ones of exposure and risks that we take in reinsurance rather than a claims projection. As you know, and I'm sure everybody's said it before, there's just very little to extrapolate right now. As a reminder, domestically, we write no personal lines in Texas. We do write commercial lines in some specialty business. Most of the exposure we have in Hurricane Harvey domestically will be through our core commercial lines, business interruption and flood, and inland marine. We have a, I wouldn't say a huge book, but a sizable enough book in inland marine in southern Texas that will be exposed to this. We have about 1.5% market share in those affected areas.
With Chaucer will be exposed to Hurricane Harvey through its treaty book, through its energy book, and through its marine book. Because the direct market, like I just said, doesn't have an estimate, clearly the reinsurance market doesn't yet have an estimate. We're both very well protected on reinsurance. Our reinsurance programs are documented in our public filings. Ours domestically is $200 excess of $900, well-placed with high-quality reinsurers. Chaucer has a reinsurance program on its inwards reinsurance business that has a retention of $25 million per cat occurrence. Just laying out some broad parameters so you can start to box this thing and understand what's exposed. As we said last quarter, we're going to have the discipline this quarter of, like many other carriers, of going out when we know to the investors when we have a good estimate.
In advance of the quarterly release, you can have whatever we know. Does that help?
No. It's funny because there's been a lot of concern with companies about asking Hurricane Harvey questions. I think that really gives us the right framing for what we can and can't know at this point in time.
Yeah, exactly. We want to hold our comments more to exposure and risk. We avoided those questions in the early morning meetings because we wanted to do it in an FD-compliant environment. That's what we're willing to say at this point. As we know more, we'll certainly tell the market because it's obviously a significant event. The last thing I would say, because I can always tell when you're looking at me is I forgot something. This is clearly an earnings event, not a capital event.
Right.
I can't speak for the industry. I can only speak for our company. It's an earnings event one time, not a capital event where we're worried about any of the issues associated with capital management ratings or anything like.
Right. Well, Hurricane Harvey may not be the last hurricane of the season.
There's another coming.
I take your point. You touched on this a little bit, and I don't know if I want to ask this specific to workers' comp or in general, we've had benign loss cost inflation, low overall inflation for a while. With regard to the antennae that you mentioned before, what analyses can you do to anticipate a change in loss cost inflation before it's really painful?
I don't have any clever answers for you because I think every company would say the same thing. I think it's how good you are at it. I will tell you that I've been unpacking loss ratios for 35 years. It doesn't matter whether it's an MLR in healthcare, a combined ratio in property casualty, or a long-tail disability income reserve. You're constantly unpacking your underwriting results to understand the attribution of the levers you pull to make money in this business with what's going on in the business. It's about veracity and velocity of information. How fast do you get the information, how insightful it is.
I will tell you that Hanover when I came into the business and started analyzing it, I think is best in class at unpacking its underwriting results and looking at underwriting initiatives and knowing that we can shave 20 basis points off a trend by doing this. Pricing segmentation. Taking that four points of rate and blasting into the market is the wrong way to do it. Giving up rate on the profitable accounts and moving rate up on the unprofitable accounts and knowing exactly where to put your pricing is both an art and a science. I would just say that we look at our loss trends in very granular fashion by geography, by coverage type. The information is quick to come out, it's insightful, and it is surgically analyzed by the underwriting teams, the claims teams in order to glean these trends.
I wasn't here at the time, but when the whole industry went through that commercial auto crisis of 2011, 2012, and 2013, it hit everybody, but I think with hindsight, you can say that Hanover recognized it first and early and jumped on it first and early. I think it's testimony to Jack Roche and Dick Lavey and the team we have, but also testimony to the systems they built to try to get ahead of this stuff before it's too late.
No, that's a great example because it's empirically true. You still have companies as recently as second quarter recognizing that there's a commercial auto issue out there. I want to shift gears a little bit to the investment portfolio, but again, if there are questions on any component of underwriting or premium growth.
How about your expense ratios? Are you pretty happy with the trend on your internal expense ratio?
Not up to now, but I will be over the next two years. Look, it was too high. I didn't come in here to financially engineer the business. I came in here to help the team grow it. Early on, I just knew the cost structure had moved ahead of the mandate we had to be high touch, narrow and deep distribution strategy, and high touch. It's the hallmark of the company. I didn't want to destroy that. Set the strategy, but I knew we needed to get at it. It was a question of when. On Investor Day, because the analysis hadn't been done yet, we only explained to you the fixed cost leverage we're going to get out of the business, 150 basis points over five years.
We're not going to wish it to happen. It's not going to happen as a matter of gravitational pull or force. It's going to happen because we managed to it. Not allowing variable costs to grow due to rate, not allowing variable costs only to grow for premium taxes and commissions, making sure we leverage our overhead, zero overhead growth in direct support. We didn't have done our work yet, but we're in the process of doing it. As we said in announce, we found $50 million of reasonably low-hanging fruit, spans and layers, individual contributors, redundant functions, under-scaled functions, three steps away from the customer. The way to think about it is if we said the expense ratio was growing at a slope like this, we said we're going to take it to a slope like this.
Think of taking a notch out of that and now growing it slower off of a lower base. I'm not sure that $50 million is in addition to the 150 basis points or part of it, but I would say it's at least a cushion against the achievement of the expense leverage we said we're going to get. The answer is no, I'm not happy with the expense ratio, which is why we took the action, and we're going to continue to look for opportunities.
Let me actually continue with that theme just for a second. I'm going to ask this question in an oversimplified way. How much of that $50 million of expense savings will be passed on in terms of lower prices? Because there is a premium growth goal, how much will fall to the bottom line?
Right now, the way we're looking at it is, first of all, it's the 118 run rate number. Little of it is happening over the next quarter or two, very little of it falls through to 2017. It's a 2018 issue. We will reinvest a modest portion of that in some of our innovation activities. The rest will be available for P&L or growth. That's not a Joseph Zubretsky sitting in my office call. That's a field call. Where do I need it? If an operator in the field comes up and says, "I can double the size of this book in five years, give me some rate," we'll do it. We don't make macro calls. We make geographic by geographic calls. If somebody needs rate and we're willing to take that bet on a clever operator, we'll do it.
I would say right now, think of it as dropping through, except for the modest component that will be reinvested in innovation.
Okay. Let me turn now to the investment portfolio. I have some statistics here. I'll read them. Net investment income up 4% year-over-year. Interest rates aren't helping a whole lot. Average invested assets up 6%, and there's a 30%+ growth rate in the other investment category. Can you help us understand what's in that particular-
Sure
asset class?
Sure. Our strategy in the investment portfolio is for it to always be there and be a meaningful contributor to our sort of the investment component of our DuPont analysis, but to never be part of the story. I think that's what you'd want to see. We're very responsible in the risk we take, the capital we allocate. We don't take duration bets. A lot of companies stay short, go long. I think it's a mistake. We manage for a complete balance between total return and book yield. I think managing for total return and not having a lot of it drop through NI isn't helpful to guys in your business and to the marketplace that tries to see the results manifest themselves in earnings. The best risk strategy you could ever have is to stay diversified. Think of our investment managers as enhanced indexers.
They're not taking wild bets and saying, "I'm going to rotate out of European financials. I'm going to rotate into pharmaceuticals." They're staying true to the Lehman Agg or whatever index you want to use. The increase in alternatives or risk assets is really a function of moving up the chain in terms of how much of our capital base we're moving into alternatives. Durationally matched on your liabilities, never waver. Our strategy is to have up to 50% of our capital base because there's no durational impact to capital really going concern. 50% of that can be in alternatives, and that's been inching up. The real increase in our net investment income is a function of the growing size of the portfolio due to our growth. That's really the story. Interest rates are not helping, but the portfolio yield of 3.4% is staying pretty steady.
Our new money yield is probably still just slightly over three. Hopefully over time, that'll ease so the portfolio rate can maintain. Basically, the growth has been in the growth in the portfolio due to our growth in premium. You're never going to see us announce some kind of exotic or leverage risk strategy to investments. We want it to be a meaningful part of the story, but never be the story.
Okay. Should we expect limited turnover in the alternative investment category because of that? Am I interpreting that correctly?
Yeah. These are not trading portfolios. There's a mortgage loan portfolio that I think that Life handles for us. There's a very alternatives portfolio that is hedge fund-like, but very low beta. These are not managed for trading and turnover. The frustrating thing, a part of the accountants tell me that a lot of the volatility actually goes through earnings, which drives me crazy, but that's just the way it is. I think it's just responsible to have 50% of your capital in responsibly capitalized risk assets that produce Book yield. If you're in stocks, and we basically invest in high-yield stocks, the book yield is actually better than bonds.
Right.
Booking a bond yield with a stock and then getting the growth in the equity isn't a bad strategy, perhaps with capital.
I was going to say we're pretty forgiving when it comes to alternative assets, but I'm not sure that that's actually true.
No.
We're pretty forgiving on the outperformance.
I understand. I've been on the other side of that.
This is an unfortunate timing question, given how weak all of P&C was yesterday.
Yeah.
I think stung you guys a little bit also.
Yeah.
If I had to guess, probably on the Chaucer side, who knows?
At, let's say current valuations, how do you think about the balance between M&A, share repurchases, retaining capital for organic growth? That is a balance process, I was hoping you could take us through your thoughts.
Absolutely. We have a very well-thought-out and methodical capital strategy. Right now, with our currency value where it is, it's hard to justify a premium-to-book value acquisition. Most of the stuff with a control premium would be too expensive, and the dilutive effect would be noticeable, if not significant. Doesn't mean we're out of the game forever. I've got a crackerjack M&A team, and when we can execute, we'll execute. What you're likely to see is what I referred to before, what I call pseudo M&A. Renewal rights deals, MGA conversions, those things are out there. Mega consolidations with agents. We're going to try to grow the book through what I call pseudo M&A. When it comes to the, what I call maintenance capital, the routine deployment of capital, here's the way we think about it.
If our ROE stays where it is, and our growth rate stays where it is, even after paying an $80 million dividend and funding the growth, you're going to have excess capital. The first thing you ask yourself, did I deploy the capital in the market? Meaning, did I write enough business on it? The answer should be yes, otherwise, we would've written more. Is there an acquisition opportunity that's accretive? Yes or no? That's opportunistic, and generally, the answer is no. Those are episodic. You ask yourself the question, do I increase the dividend, pay you a special dividend, or buy back shares?
Most investors tell us, for that amount of excess capital, $50, $70 million a year, we love the earnings per share accretion we get when you buy back shares and let it flow through to EPS, so that the shares of THG grow in my portfolio instead of getting this weird dividend. It's not enough to pay a dividend anyway, but share repurchases and the maintenance capital is the way we think about it. If you look at our ROE and 7% premium growth, our premium to surplus ratio, and how much capital we need to reserve, it still throws off excess capital over the planning horizon, and we would repurchase shares on a opportunistic basis to give you the EPS lift. That's the way we think about it.
Yeah. Fantastic. We have time for one last question, so I wanted to see if there was any interest from folks in the room. In the absence of that, let me ask, it's sort of an oddball question, but it goes to at least how we think about valuation. If you had to pick between growing book value and growing returns on equity-
Yeah
How do you make that decision?
They're so hand in glove because I've been asked that all the time. They're so hand in glove. I think having the capital cushion, your earnings cushion of an ROE is always more comforting. We've always said that we would sacrifice growth for margin. Whether when it was in other businesses or this business, I think you'd always opt to have $1 of margin instead of the growth. I probably would drive for the combined ratio and the ROE more than I would for the growth in book value. At the end of the day, I would look at you and say, I'm not going to ask you a question at your own conference, what is the trade-off?
When we look at the multiple of book value that the ROE is going to produce and the actual book value, where's the inflection point of where you'd make that trade? It's really hard to discern. I'm telling you right now, putting that aside, if we can get close to $100 of book value and a double-digit ROE, which we said at the low end would be 11, at the high end 12, I'm not sure there's a bad story within that construct at all. That's the way I would answer it. Your question is an interesting one, and it's really hard to parse it that granularly to give you a hard conclusion on that.
Right. I'll answer your question anyway.
Please.
I've probably told 10,000 people over the last 15 years that book value doesn't matter. It's all about the earnings. I don't think I've convinced anybody in that period of time, I probably won't start now.
Well, like I said, at the end of the day, you're in the insurance business, things can turn, having an extra few dollars of margin to play with-
Yeah
is always, since I was this high in the industry, was taught that that's the mindset you need to have as you're managing these businesses. I'm glad we agree.
Excellent. Anyway, please join me in thanking John. This was phenomenally informative.