Good day, everyone. Welcome to Selective Insurance Group's Second Quarter 2019 Earnings Call. At this time, for the opening remarks and introduction, I would like to turn the call over to Senior Vice President, Investor Relations and Treasurer, Rohan Pai.

Good morning, everyone. This call is being simulcast on our website, and the replay will be available through September 3, 2019. A supplemental investor package, which includes GAAP reconciliations of non-GAAP financial measures referred to on this call, is available on the Investors page of our website, www.selective.com. Certain GAAP financial measures will be stated in the call that are also included in our previously filed annual report on Form 10-K and quarterly Form 10-Q reports. To analyze trends in our operations, we use non-GAAP operating income, which is net income excluding the after-tax impact of net realized gains or losses on investments, unrealized gains or losses on equity securities and debt retirement costs related to our early redemption of our debt securities in the first quarter. We believe that providing this non-GAAP measure makes it easier for investors to evaluate our insurance business. As a reminder, some of the statements and projections made during this call are forward-looking statements as defined by the Private Securities Litigation Reform Act of 1995. Forward-looking statements are not guarantees of future performance and are subject to risk and uncertainties. We refer you to Selective's Annual Report on Form 10-K and any subsequent Form 10-Qs filed with the U.S. Securities and Exchange Commission for a detailed discussion of these risks and uncertainties. Please note that Selective undertakes no obligation to update or revise any forward-looking statements. Joining today on the call are the following members of Selective's executive management team: Greg Murphy, Chief Executive Officer; John Marchioni, President and Chief Operating Officer; and Mark Wilcox, Chief Financial Officer. And with that, I'll turn the call over to Greg.

Thank you, Rohan, and good morning. I'll first make some introductory remarks and then focus on some high-level themes and initiatives that enhance our strategy and position us for continued profitable growth. Mark, then will discuss our financial results, and John will review our insurance operations in more detail, providing additional color on key underwriting initiatives. Our second quarter results were excellent, reflecting ongoing superb underwriting results coupled with outstanding investment income performance. For the quarter, our combined ratio was 93.1% and after-tax net investment income was up 27% to $48 million that generated non-GAAP fully diluted operating earnings per share of $1.16. For the first half of the year, we generated very strong annualized non-GAAP return on equity or ROE of 12.8%, which exceeded our full year target of 12%. Our ROE targets are established annually based on expected interest rate levels, our weighted average cost of capital and excess margin over weighted average cost of capital, as well as the general property and casualty market conditions. The targeted ROE then determines our insurance product pricing on a risk adjusted basis by Line of Business and provides the baseline for the financial portion of our annual incentive compensation plan. For the quarter, all aspects of our underwriting operations performed remarkably with: one, overall net premium written growth of 7%; two, solid renewal pure price increases of 3.4%; three, strong retention; and four, new business that was up 7% to $146 million. We continue to execute on our initiatives around increasing our share wallet within our Ivy League distribution partners appointing new partners as well as growing in our five recently opened states. Based on the first six months, our insurance operations generated an annualized ROE of 6.4 points. We continue to maintain a leadership position at commercial lines renewal…

Great. Thank you, Greg, and good morning. For the quarter, we reported a record $1.21 of fully diluted earnings per share and $1.16 of non-GAAP operating earnings per share. We generated an annualized ROE of 14.5% and a non-GAAP operating ROE of 13.9%. Through the first six months our non-GAAP operating ROE of 12.8% is above our 2019 12% ROE target. We are pleased with our track record of generating consistent double-digit ROEs for the five consecutive years. For the quarter, our high quality underwriting results contributed 7.1 points of ROE and investment income contributed a solid 9.6 points of the overall ROE. We enjoyed another strong quarter of growth with consolidated net premiums written up 7%. Underwriting profitability was strong with the second quarter combined ratio of 93.1%. On an underlying basis or excluding catastrophe losses and prior year casualty reserve development, our combined ratio was 91.1%. For the first half of the year, our reported combined ratio was a profitable 93.9% and our underlying combined ratio was 92.1%, which represents a 160 basis points of margin improvement over the strong yield -- over the first half of 2018. Our 90% ex-cat combined ratio for the first 6 months is better than our original full year guidance of 92% for 2019, and as a result, we have updated our full year ex-cat combined ratio forecast to 91% for the year, assuming no additional prior year casualty reserve development. For the quarter, catastrophe losses added 4.6 points to the combined ratio, which was in line with our expectations, which is seasonally adjusted. Non-cat property losses accounted for 14.4 percentage points on the combined ratio, which is slightly better than expected. In addition, in the second quarter, we experienced $17 million of net favorable prior year casualty reserve development, driven by…

Thanks, Mark, and good morning. I'll begin with the results of our operations by segment and then provide an overview of some of our strategic initiatives. Our Standard Commercial Line segment, which represents approximately 80% of premiums, generated net premiums written growth of 8% for the first half of the year, continuing its track record of strong profitability. For the six months, the segment generated new business growth of 10%, stable retention of 83%, renewal pure price increases of 3.2% and an excellent combined ratio of 93.7% or 92.7% on an underlying basis. Our highest quality standard commercial lines accounts, which represented 49% of commercial lines premium, we achieved renewal pure rate of 1.8% and point of renewal retention of 91%. On the lower quality accounts, which represented 11% of premium, we achieved renewal pure rate of 8%, while retaining 78% at point of renewal. Our ability to analyze the risk and return characteristics of each piece of business at an extremely granular level allows us to achieve additional loss ratio improvement through mix of business changes, while maximizing overall retention. Going down to the results for the first half of the year by commercial line of business, our largest line general liability achieved an 87.8% combined ratio, which included favorable reserve development, totaling $7 million, or 2.1 points. We achieve renewal pure price increases of 2.4% for this line. While loss trends have been generally benign in recent years, we are closely monitoring loss severities. Our workers' compensation line generated an 83.9% combined ratio for the first half of the year, aided by favorable reserve development, totaling $20 million accounted for 12.7 points on the combined ratio. This favorable development related primarily to lower than expected severities for accident years 2017 and prior. Renewal pure pricing was down 2.8%…

Certainly, we will now begin the question-and-answer session. [Operator Instructions]. Our first question is coming from the line of Amit Kumar of Buckingham Research. Your line is open.

Good morning, Amit.

Hey can you hear me?

Now we can thanks.

Sorry about that.

Now that is fine, good morning.

Just very quickly on some pricing. You talked about the industry facing the need to pursue pricing because of past issues. Would that translate in your pure pricing modestly declining going forward as you look at this market share or could there be some pressure from the new business penalty for you?

Yes. So let me start and John can certainly head in. So -- as I went through our whole presentation -- our pricing is one of the most disciplined processes that we're going through. And as I mentioned on the call, we're looking at the past four-year odd level, that's your baseline, where do you start, what's in that? What level of noises in that and how do you expect trend to move forward? And then where is your investment returns? So as I mentioned the three factors, right now, the one, that we expect the most amount of pressure to be honest maybe on the investment performance due to one, the lower, where we are on the LIBOR in the 90-day LIBOR, where we are in the US 10-year and the fact that spreads have narrowed in roughly in the 40 basis point range across many sectors. So that part of it is going to put additional pressure to put pricing at the right level for 2020 right out of the box. I would tell you that what I had in my prepared remarks is that the favorable aspect that we view is one, the fact that our current observed levels of current accident year claim count for casualty is below what we expect. So that is your early indicator of how your current accident year is going to move, because there's two things that affect your accident year; there's frequency and severity. And if your frequency is off the mark early on, you know that you're going to have pressure on your pure premium calculations. So we're not seeing that. And the other part then that is relieving a little bit of pressure is the fact that we continue to see favorable development and the favorable development is basically lowering that starting point that we're at and why we are kind of favorable on where we stand as we feel all of those things line up well for us. But the pressure is probably more applied on the investment income side that it may be is anything underwriting specific.

No. I think Greg covered a lot there. Amit, this is John, and let me just add a couple of additional points as clarification here. So because we're achieving our target margins, our focus from a pricing perspective and how we set our targets is focused on maintaining those current profit levels. To the extent as Greg indicated investment returns come under increasing pressure as we look out to the future quarters that requires us to lower our combined ratio expectation to make up for the difference from an overall ROE perspective and that's how we'll manage it going forward. Now as you heard from Greg's earlier comments there is pressure in the market to improve underwriting results. And I think this is also part of your question, that does present opportunities for us. Because our margins are strong and we have -- and our views now for well over a decade, very sophisticated underwriting and pricing tools as companies start to take more aggressive stance on rate. They want to pushing some high quality accounts into the market that present great opportunities for us to write new business. And we do think that's a benefit to us in terms of how our position. The other point I think worth noting is we also strive to make sure that all of our lines of -- major lines of business are either achieving or approaching their target combined ratios. And as you've seen in our results and our prepared comments, we are still working to lower commercial auto and part of that is through rate, part of that through other underwriting mix in claims, improvements. And property while generating a positive underwriting result is running above where most companies would want it to, because of the inherent volatility in that line. So these lines of business for the industry are being masked somewhat by the very strong results in workers' comp. But workers' comp now is in a negative rate situation and when you look out to the future that's going to be addressed. So there's a number of pieces there, but it's certainly a dynamic marketplace.

So, if I step back, do you get the sense on the commercial piece and this is maybe a broader question. Can you talk about what the loss trends might be running at in terms of some numerical values? I'm trying to figure out how much will pricing -- how much is pricing versus loss cost trend margin running at?

So in our commentary we had that we believe it's in the 3% to 4% range and that's --

The rent overall?

Yes. So loss trends overall is what we feel we're running out today.

So still we get complaint. Let's just make sure we make this point. I know you guys are hearing a lot from our competition and -- so let's just baseline for a minute, if we could. So our pricing in commercial lines running in the 3 spot 2; 3 spot 3 area. If you looked at it all in, on a renewal basis like many of our competitors are reporting, which I'm not a believer in were like 2x. So I just want to make sure that when you guys sit there and you start to compare, Murphy told loss trend was going to be in the 3x to 4x renewal pricing is a 3 spot 3, I've got a margin diminishment, all right. That's where we are pure pricing. I just want to make sure that that is clear for you guys, one. Two, then, and John has talked about this over and over again. We drive the hardest a lot out of our underwriting and claim improvement based on what we're doing in terms of the segmentation of our business and how we are not kind of driving higher rate level or non-renewing some of the accounts at the poorest performing levels and that's how and plus all the claim activity that we've done, that has improved the cost of goods sold, all work in as part of our overall improvement. And again I want to -- I don't want to sit here and tell, hey we're up 6% in overall renewal pricing, but because I don't believe in that, but I just want to make sure that you guys have all of the moving parts when you start to compare Selective and Selective's performance for 2020 versus some of the comps you may be looking at. That's all.

Yes. That's another to verify your comment because everyone has a different way to compute what the pricing number is, and it's not apples-to-apples in many cases. So --

That's the point that I'm trying to make. And when we start reporting results ex-comp, and then compare the ex-comp results to your rate level, ex-comp and then tell me how those numbers stack up.

Yes. That's a fair comment. But the only other question, and I will stop after that is, it's interesting to listen to your comments in commercial auto, when you said it was -- it seem that it was stable versus Q1 and I feel like commercial auto is that Elephant and everyone has a slightly different take on it. Can you just talk about what changed Q2 versus Q1 and also maybe talk about the broader discussion as it relates to commercial auto and some other lines regarding the jury awards and the total environment and has the thought process changed on that or not? Thanks.

So let me start, and then John -- this is Greg, so let me start. First of all, there is no change in our view at Q2 versus Q1. Our current accident year is running at above six. It stays at above 6. I think the commentary that you're hearing is just more data accumulated. After the first quarter, we told you that overall observed trends were better than anticipated on the claim count casualty side and that was for the overall book and I don't want to get down to such specificity by line, but that overall trend continues and that's our view of commercial lines in totality. But the commercial auto has our view at that six months versus three months is not different. What is the benefit is some of the pressure that we've seen on the prior accident years, we're not seeing any of that yet today and view that we're comfortable with where we are reserved well levels where we have our severity pits for the 2018, 2019 and 2017 years, feel good about that and feel good about [indiscernible]. But I think where people got offline or a little bit off their budgets, which just principally driven by claim frequency counts and then followed by some elevated severity, which is what you're talking about, we have seen some pockets of higher severity and certain parts of the country, that's something that we are closely monitoring, it's not an overall trend. But you know, we are very, and much tuned into what's happening on the severity side of the book. So let me turn it over to John to add some additional color to that.

Let me just address the second part of your question relative to additional trends, social trends. We do monitor litigation rates and attorney involvement rates and across all of our casualty lines including auto, they have been very stable. Now we certainly hear other market participants, public commentaries on that topic. And we're mindful of that. But from our perspective we have not seen any material change in litigation rates. But as I've also referenced in the past, we really pride ourselves in our claims organization in early communication with claimants, an ongoing communication with claimants, which I think is the best way to give them a sense of how the claim is going to be adjudicated and in many ways will reduce the likelihood for them to feel the need to involve an attorney in that claim adjudication.

Your next question comes from the line of Paul Newsome from Sandler O'Neill. Your line is open.

I just wanted to -- not really beat the commercial auto, the dead horse, but actually I want to ask another question that maybe thinking you bring all that. I have a theory that there is a difference between folks that use ISO only policy forms in rates versus those that tend and use their own system and own pricing. So my question is, just as my theory is certain with what you've experience and where do you sit on that spectrum of how you underwrite your commercial auto? Is it mostly your forms and data, or is this a lot of ISO and industry level stuff?

So, Paul, this is John. Let me take a crack at this. We are largely an ISO-based company, in terms of forms, rules and rates. We do have our own proprietary endorsements and those are used to modify coverage in many cases they add coverages that are segment specific for different industry classifications. But I think the important point, and our sense is that's largely the case for most market participants, whereas in personal auto, it's a little bit different where some of the bigger companies are using their own underlying data and are less relying on ISO. For commercial auto it's predominantly ISO-driven. But I think that's really just the underlying rate plan that starts the process in terms of pricing adequacy on an account-by-account basis. What's equally important is the accuracy of the information you have relative to vehicle type, vehicle radius, vehicle usage, that's an important consideration, but also the underlying pricing models that we use. So when you think about the ISO class plan that's what underlies your base rate for the average risk by certain classification type. You then modify that based on individual risk characteristics in our case, using our multivariate predictive model that separates from a future profitability expectation best from worse and modifies -- recommends to the underwriter how to modify that file base rate, which is based on ISO loss costs. So that's our approach. We think it gives us an advantage. We're not alone and taking that approach, but there is a lot of companies in the market that are just relying a 100% on the ISO class plan not using their own tools and models to modify it.

Could you talk a little bit more about the E&S business turnaround that was quite dramatic? And what are your thoughts sort of prospectively, now that you seem to have, you've got that to a really decent level of profitability?

So I'll start again. Paul, this is John. We're pleased with what we're seeing there. And again it's been a few quarters in a row now and we like that. We're not declaring victory by any stretch, but we think we're on a very good path. We have seen very strong property results that underlie the results that you're seeing over the last few quarters that we'll have inherent volatility in it going forward just like our core commercial lines business as underlying volatility on the property line of business. That's less than a quarter] of our premium, or about a quarter of our premium [indiscernible] asked we're predominantly a casualty writer. There has been no prior year development, unfavorable or favorable in the last couple of quarters, which is also a positive. So you're getting a sense of what the underlying casualty book is running and you have seen improvement. That improvement is driven by pricing actions that have been taken, it's been driven by underwriting actions in particular exiting a few smaller segments of the business that we're not generating significant premiums, but we're adding some volatility to the results. And also a lot of the claims changes that we made by migrating to our Selective claims operations have improved the performance of that portfolio as well. So we feel good about where it is. We continue to believe and this has always been our philosophy getting into this business is, we expect to achieve our target margins for this business consistently and we're willing to accept a little bit more volatility quarter-to-quarter, year-to-year with regards to premium growth based on market conditions. As we sit here today, we're getting solid growth. We've got solid margins coming through, and we're pleased with the path that we're on. But to the extent the pricing starts to drop off, but we don't believe we're going to achieve our target margins will at the top line flattened out and that's the case.

So Paul, this is Greg. Just that only add is that, so we want to get the contract in binding business growing again. We'd like to see a little bit more discipline in the contract binding pricing and that's where I would say that what we experience doesn't necessarily match up with the broader base E&S trends that you are hearing about margin-wise, but again like John said, we feel good about where we are on the operation and want to continue to maintain performance at this level, but it is a small book of business and the numbers can get pushed around by not that large events.

We have a question coming from the line of Christopher Campbell of KBW. Your line is open.

I guess, first question is on personal auto, so rates accelerated and premiums were down year-over-year, I guess how would you just describe the competitive environment you're seeing in personal auto?

Chris, I think clearly what we're seeing is a rapid shift down in rates and that doesn't necessarily mean negative, but generally speaking broadly across the market, what was a pretty strong rate environment and based on the rate filings we see come through company-by-company state-by-state has quickly dropped down into the very low-single digits and again generalizing each company is going to be a little bit different, but it has moved fairly quickly, and I think it's in reaction to some of the bigger companies that I've seen improvement in our own underlying trends, but that's really what happens. We've continued to have additional rate running through our book relative to the filings that we've made over the last couple of quarters and that's really hurt our competitive positioning for new business based on the rapid change in the market environment.

So what's your view, I mean, if you're thinking competitors are going to low single digits what's your view of like loss cost inflation for the industry?

We don't view loss trend in personal auto really that different than overall. So I'd put it in the 3% to 4% category until you start to see miles driven our gas prices, or other things really start to fluctuate right now with very low unemployment there is a lot of people on the road. And let me just add one other comment, Chris, relative, I mean we're trying to improve. We feel really good. We've always viewed personal lines as three different operations within that division; the home, and as we've always told you our goal in home is to get it into a low-90 in a normal cat year, normal cat year in homes about 14 points. And we're pretty much there when you look at it on a multi-year basis. We have two states that we continue to focus on the home, but on the home book that we need to make further improvement. Our flood operation improves our combined ratio by about 60 basis points and it's the one hedge that we do have to adverse weather. So that's a positive because we actually made additional funds based on how the claim reimbursement works and the FEMA program and that's a positive. And then the auto, which is what John, was touching on. You're stuck in the comparative rates, the question is we need to make some fine-tuning improvements in our expense ratio. And then I think, look at how we militate some of the increases in the overall and target some of our increases that we need to do very granularly so we improve our positioning in that comparative greater that's used in every agency.

And then kind of just 1 more like on the commercial line side. So these competitors retrench and start to raise rates. How does this impact like your willingness to grow? So I would think that your combined -- your core combined ratios are pretty good, right. I don't think you guys really need to take it on a rate. So, I mean, but you still look like mid single-digit rating or premium increases. I mean, can you get the double-digits in commercial lines like if everybody starts raising rates?

Yes. As I said in my commentary, again it's some of it depends on how the sophistication of the competition does it. I would say to you today anybody that socialized rates going to get into hurt badly, because if you only have -- if you don't have the sophistication that John walked you through just in commercial auto and therefore your distributing rate more socialized across your segmentations, across your accounts, across your geo-locations, that's a huge opportunity because those accounts will enter the market and then we will have an opportunity to run our very sophisticated modeling against and see where we price out at, and I view that as a competitive advantage in this marketplace. And this is the opportunity that we would want to grow. And if we had the opportunity to post some out-sized growth levels, we would be willing to do that, but it has to be at the right pricing in the marketplace.

Yes. Just add to that, Chris, to Greg's point, to the extent we have opportunities to put our foot down and grow a little bit more rapidly. In the past, we've talked about the sustainable growth rate probably about 75% of the forward ROE, given the dividend payout ratio. And let's call it around 9%. We grown at about 7% thus far year-to-date. We're at the low-end of our range, as I mentioned in my prepared comments in terms of a premium to surplus ratio and we have adequate and ample capacity at the whole curve to drop down into the insurance subs. So from a capital position, from a leverage perspective, we have ample capacity to grow market opportunities present themselves.

We also have a question from Mike Zaremski of Credit Suisse. Your line is open.

Greg, in your prepared remarks, you said you feel pricing for the industry, commercial wise should probably stay around current levels and you talked about lower interest rates. And I also believe you; you talked one of the reasons being some competitors seeking to improve their results. I didn't -- I might have -- so I'm just trying to -- if I'm understanding you correctly, just want to understand -- do you feel that a number of your competitors are hurting? And if so, is it commercial auto, like, is something changed in the last quarter or two, or you feel kind of more confidence that the industry is -- you feel more confident about your views on pricing?

Well, let me just start with the fact that our expectation is different than a lot of other companies. So let me just make sure we've got that clear. I mean, our core baseline is, we want to print a number in excess of our weighted average cost of capital. That's where we feel we need to be. We told you last year that when we started 2019, our target is 12. We will recalculate that target for 2020, and we'll tell you what that target is and that becomes the basis for what we need get rate wise and it becomes the basis for our incentive compensation plan. So those are all very much in line and where the rest of the competition is, I mean, you can tell me, you look at the performance, it's all over the place. There are some very large commercial lines writers that are not printing margins even at the weighted average cost of capital. And so what I'm really referring to is the fact that a) not a lot of companies are at the baseline they need to be number one. Number two, as the low interest rate comes in, companies are going to have to make that up in improvements and the question is, if you thought you needed a certain amount of improvement, now you need more than that because of the lower yield environment. And now the question is, does the competition have the tools, agility, sophistication, to be able to deploy that added a granular level? And when they don't have that capability, we are very sophisticated now in our ability to go out and harvest that business to increase our share of wallet. And these are the things that John is mentioning as we try to drive…

And so you've talked a lot about telematics on the commercial side over the last couple of quarters or maybe more than that. So do you -- are you guys feel a first mover in that, in telematics for commercial auto? So I think only you guys progressed right now and talk about it.

We think we're the first mover in terms of how we've deployed it. Again it's telematics-based we're not using it for rating purposes like you see on the personal line side. This is more a way to give a value-added offering to commercial business owners. So they could better manage their fleets. And at the same time, because of the Telematics you're creating greater awareness around driving behaviors of individual drivers in these fleets of vehicles, which we think improves experience overall and gives agents some additional service to offer their clients, which will help them win and keep more business. As far as we've seen there is nobody else that's deployed a product like that, but it's not the same telematics-based rating platforms you've seen really permeated through the personal line side of the business.

So, Mike, this is Greg. And when you think about, so obviously there's three things that we do as an industry: one, we're a form of capital, so by an insured -- having insurance they need less capital on their balance sheet; two, it provides back together after a claim; and three, we make our community safer. One of the biggest things that our drive product does and we are pushing it hard is because it's not plugged into the computer, the vehicle works through other power sources, is actually works off the phone. And because it works off the phone, it can detect distracted driving, which is a huge issue in a marketplace. Selective Driving is something that we view as a major issue something we want our trusted Risk Advisors, our agents, our producers pushing at point of sale. That look, we want to make you a safer account by having the Selective Drive product not only do you know where your vehicles are, not only do you get a scoring on every one of your drivers for every week, how well they're driving, but you also understand when they distracted driving in addition to fleet maintenance, an addition to other things that you want to know as an owner. So that's a valuable tool to an owner, particularly we've had a lot of construction business for an owner of the construction account to know where all other people are, know where all other vehicles are. On a Friday afternoon at 2 o'clock, you want to know who is still on the job site versus who is not on the job site. So this is what you get from this product and it's offered as part of just being a Selective customer. So again it's a real value-added product that is very unique in the marketplace.

I'm sold in the product. I just don't --

[Indiscernible] commercial lines account, I'll have the drive account this afternoon. I personally deliver to you, Mike.

Last question might be for John. To the extent it's not proprietary. Maybe you can give us a flavor some of the strategies, carriers like yourself used to kind of help mitigate workers' comp suggested rate decreases. Lot of times, we see some of these headlines from certain state saying 10%, 18% rate decreases and I -- I know, I'm curious like yourself talked to being able to mitigate some of that.

Yes. So let me just be clear on how we manage pricing. We reprice individual accounts based on the pricing tools that we have relative to the right price for the exposure presented. Loss cost filings do change the headline number that you see relative to an individual states loss cost change is the overall book in the state that it applies to, it affects each individual book differently based on your mix of industry classifications, so it's not a straight comparison. There are individuals credits and debits on individual accounts that are based on risk characteristics. Those will move on a renewal policy based on the underwriter's review of the change in characteristics, as well as the change in the underlying loss cost. And that's how we administer pricing on an overall basis. It has resulted in pricing. In our actual renewal portfolio, it's different from what you see on the overall loss cost filings, but it's still our focus is always on making sure that each account is priced appropriately based on the exposure presented.

So if there is -- if the industry is in a credit position, you can give them less current less credits and that offsets some of the mitigate some of the suggest.

Yes. Mechanically that's how it works. But again the underwriting documentation behind that needs to explain why that's appropriate for that individual account, and that's what we really pride ourselves on.

And Mike, we got a great claim operation on the comp side. That's very focused in on pure premium or cost of goods sold whatever words you want to use on it. So they are really doing down in terms of tendencies when we get a claim, how do we get it fast track. Are we creating a claim, a climate for a back injury or obesity or both? And then how do you manage the outcome on that claim. How do you escalate them to the right people and we're actually now looking at our cost of goods sold separated our pure premium calculations separated into very granular level, whether it's needs back shoulders, okay, what's the need costs. How many needs you're doing? What are we seeing in the trend in needs? And then how is that driving overall your cost higher. This is what John is talking about. You just -- right now it almost seems like the NCCI calculations have frequency going 0. And that's not going to happen, you're going to see an uptick in frequency, and at some point you have to be very mindful of severity in this line and when it turns, it's going to turns, we think it's going to turn pretty severely, and that's why we're very careful about it, and that's why we see that are very disciplined in what we're trying to do in the marketplace. We're running on an accident year basis, our comp at about 96% right now. On an accident year basis, that's where we are, but that's not that far off what our target risk adjusted performance should be. So is there a little bit of room in there, yes. But part of it may be affected by prior years continuing to drop down, but still at the end of the day, a 96% versus your target, you don't have enormous amounts of room in that today.

But just 1 follow-up, you said NCCI's frequency going to 0. If I understand --

It's a joke. That's a joke. They've got frequency in maturity going to zero registry wide. It's a joke, sorry. I should have classified it as a joke, but frequency is not going to zero, Mike, okay. I'm sorry.

But the launch of training frequency is negative, right?

It has been declining, but I would say more of the story in comp, particularly in all this favorable development you're seeing has been severity focus and what your long-term inflationary embedded trends were in medical versus what you were actually experiencing that's where a lot of the release, it's been a little bit of frequency, but it's been a lot of severity buried in these numbers. You know your frequency count in comp within 18 months. I mean, I actually sit across the table shaking his head. I will tell you that percent competence that he has, but I know what it is. And so your frequency numbers really aren't moving around that launch. It's the severity that's really changing and it's changing versus what they've embedded in that diagonal they go way back to the beginning of time. I could say it that way. But I wanted to clarify for everybody; the zero frequency is a joke.

At this time, we don't have any questions on queue. Speakers you may proceed.

All right. Thank you very much. If anybody has any follow-up, please contact Mark, or Rohan. Thank you very much for your participation in the call today. Thank you.

Thank you everyone. And that concludes today's conference. Thank you all for joining. You may now disconnect.