Donegal Group, Inc. (DGICA) CEO Kevin Burke on Q2 2018 Results - Earnings Call Transcript

Donegal Group, Inc. (NASDAQ:DGICA) Q2 2018 Earnings Conference Call July 31, 2018 11:00 AM ET

Executives

Jeff Miller - Chief Financial Officer

Kevin Burke - President and Chief Executive Officer

Analysts

Christopher Campbell - KBW

Bob Farnam - Boenning and Scattergood

Operator

Good morning. My name is Jessie, and I'll be your conference operator today. At this time, I'd like to welcome everyone to the Q2 2018 Earnings Conference Call for Donegal Group. All lines have been placed on mute to prevent any background noise. After the speakers remarks there will be a question-and-answer session. [Operator Instructions]

Jeff Miller, Chief Financial Officer, you may begin your conference.

Jeff Miller

Thank you very much. Good morning, and welcome to the Donegal Group conference call for the second quarter ended June 30, 2018. Yesterday afternoon, we issued a news release outlining our quarterly results. For a copy of that release, please visit the Investor Relations section of our website at donegalgroup.com.

It was an eventful quarter at Donegal Group. In a minute, Kevin Burke, President and Chief Executive Officer, will provide commentary on the quarter, discuss recent developments and update you on our business strategy and initiatives. I'll then provide a brief overview of our quarterly financial details. Following our prepared comments, we will open the line for any questions you may have.

Before we get started, you should be aware that certain statements made in our news release and in this conference call are forward looking in nature and involve a number of risks and uncertainties. Please refer to our news release for more information about forward-looking statements.

Further information on risk factors that could cause actual results to differ materially from those projected in the forward-looking statements is available in the report on Form 10-K that we submitted to the SEC. You can access our Form 10-K through the Investors section of our website under the SEC filings link. We also provided a reconciliation of non-GAAP information as required by SEC Regulation G in the news release we issued yesterday afternoon.

With that, I'll turn it over to Kevin.

Kevin Burke

Thanks, Jeff, and welcome, everyone. I think it's appropriate to start the call by highlighting two recent developments here at Donegal before we dive into the discussion on the second quarter and our strategy for the remainder of the year and beyond.

To begin, Donegal Mutual, majority voting control of Donegal Group, announced that Don Nikolaus will retire on September 1, 2018, after serving as President and CEO for the past 37 years. Don Nikolaus took the helm of Donegal Mutual Insurance Company back in 1981 and led the development of the company from a one-state mutual insurance company writing approximately 30 million in annual direct premiums to the lead company of Donegal Insurance Group, which today includes 11 insurance companies, serving 26 states and writing approximately 1 billion in annual direct premiums.

Don will continue to serve as Chairman of the Boards of Directors for Donegal Mutual Insurance Company and the subsidiaries of Donegal Mutual Insurance Company and Donegal Group Inc. and will then serve as a consultant to Donegal Mutual Insurance Company and Donegal Group Inc. and our respective Boards of Directors following his retirement.

Second, we announced that we entered into an agreement with Northwest Bancshares for the sale of Donegal Financial Services Corporation and Union Community Bank. We originally formed Donegal Financial Services Corporation in 2000 as part of our banking investment strategy. We made the decision to pursue sale of our bank earlier this year because banking is not a core business for Donegal, and the sale of this investment would allow our management team to focus entirely on the growth and profitability of our insurance operations.

We ultimately selected Northwest as the acquirer of our bank because of the excellent cultural fit between our respective organizations and because we believe Northwest will ensure seamless transition of the operations of the Union Community Bank and enhance services and products to its customers.

And finally, the pricing environment was favorable to allow us to obtain terms that would ultimately be beneficial for all of our stockholders. We currently expect to close on this transaction in the first quarter of 2019, and we'll utilize the proceeds from the sale to support our strategic goals as we focus on our core property in casualty insurance business.

With that, let's move to a discussion of our insurance operations. While we have more work to do, we made progress during the second quarter towards our goal of enhancing profitability levels of our underwriting operations. We experienced a very challenging first quarter in which we encountered weather-loss activity that was well above our historical average, a larger-than-normal incident of fire losses and significant prior year reserve development within our personal and commercial automobile segments.

During the second quarter, weather-related losses and a handful of workers' compensation losses masked underlying improvements within our core results. For example, the initiatives we have put in place to restore profitability within our automobile lines are beginning to show signs of traction. Without going through an exhaustive reiteration of what we stated in the first quarter, we began to see escalating trends of higher losses, particularly in terms of ultimate severity and delay in claims reporting. Recognizing that this has been an industry-wide challenge, we are taking aggressive steps to address these trends.

As we have said in the past, the causes vary from distracted driving, higher vehicle repair cost to ultimately a greater preponderance per plaintiffs counsels to delay the reporting of claims information that allows us to project ultimate settlement cost. We continue to be very thorough in examining each claim, defending against claims that have no merit and working to settle valid claims fairly and efficiently. We have implemented and will continue to implement automobile rate increases in all the states in which we are actively writing business. In personal auto, we have been using a predictive model in our underwriting process for years. We have tightened our underwriting guidelines and implemented a number of actions that have slowed new business growth so that we are no longer taking on additional underpriced business.

Our approach is a combination of predictive analytics, rate refinement, traditional underwriting review as well as having the discipline to accept the lower growth rate in order to restore rate adequacy. In commercial auto, we are taking very similar steps. We are in the process of extensively re-underwriting all commercial automobile renewals in several underperforming states, including Georgia, which we highlighted last quarter as the challenging state from a profit perspective. We're taking definitive underwriting and pricing actions based on scoring of each renewal policy in these states.

We are using our predictive model to score all new commercial automobile business and renewals for underwriter review. And our commercial underwriters are utilizing the model scores to determine tier usage, pricing increases and other appropriate underwriting actions. Predictive models are also now in production for BOP, CPP in contractor new business policy. While we are addressing the problem head-on through various pricing and underwriting actions, we have seen our retention rate stay relatively stable despite these actions.

And to this point, market conditions have allowed us to achieve steady net written premium gains. We expect higher premium rates and enhanced implementation of technological advancements to yield incremental profit improvement over time, as our net premiums earned reflect more appropriate pricing for the automobile risks throughout all of our marketing regions.

With that, let me turn the call over to Jeff to go through some of the other aspects of our financial results, and then I'll close out the comments with some longer-term strategy discussion.

Jeff Miller

Thanks, Kevin. I'll briefly go through the financial results and provide some additional details. Beginning with premiums, our net premiums written increased 2.7% to $195.9 million for the second quarter of 2018, compared to the prior year second quarter. The increase includes 5.8% growth in commercial lines and virtually no growth in personal lines. In commercial lines, we saw a combination of new policy growth and modest renewal premium increases on average with higher renewal premium increases in commercial auto.

In addition, the increase in other commercial lines net premiums written reflects the modification to third-party reinsurance coverage related to umbrella liability policies that was effective March 1, 2018. We are now retaining a higher amount of premiums and losses for our umbrella liability risks, which have historically been very profitable. In personal lines, the impact of rate increases was offset by net attrition as a result of higher rates and the underwriting measures we've implemented to slow new policy growth. Net premiums earned of $185.7 million for the second quarter 2018 increased 6.1% compared to the second quarter of 2017. We expect our premium growth to reflect a proportionately higher impact of rate increases versus exposure growth as the year progresses.

Moving to losses. Our second quarter 2018 loss ratio was the same as the prior year period at 73.1%. Weather-related losses contributed 9.5 percentage points to the loss ratio for the second quarter of 2018, compared to 11.5 percentage points for the second quarter 2017. Workers' compensation losses in excess of $50,000 were $7.1 million or 3.8 percentage points to the loss ratio in the second quarter 2018, compared to $3.8 million or 2.2 percentage points to the loss ratio for the second quarter 2017. The increase in severity was due to a handful of unusual losses.

Fire losses in excess of $50,000 were $6.7 million for the second quarter of 2018 or 3.6 percentage points to the loss ratio. That amount was modestly lower than the large fire losses of $7.6 million for the second quarter 2017 or 4.3 percentage points to the loss ratio. We noted a modest decrease in the impact of both homeowners and commercial fire losses in the second quarter 2018.

Development of reserves for losses incurred in prior accident years had virtually no impact on the loss ratio for the second quarter 2018, compared to 3.3 percentage points of the loss ratio for the prior year quarter. While we saw a continuation of the trends we discussed in the first quarter call related to the delay in our receipt of information on previously reported claims, the impact on both personal and commercial auto reserves was significantly lower than the first quarter 2018 amounts. We were pleased that our actuarial analysis for the first half of 2018 resulted in expectations that were largely consistent with the first quarter review and required no further prior period reserve strengthening actions.

Our actuaries did shift some prior period reserves between individual lines of business, which resulted in around $1.8 million of adverse development in personal auto. However, that development was more than offset by modest amounts of favorable development in our other commercial business lines netting out to about $500,000 in favorable development for the quarter.

Our expense ratio improved slightly to 31.8% for the second quarter 2018, compared to 32.6% for the prior year second quarter. We attribute the decrease to a reduction in underwriting-based incentive costs for the second quarter 2018, compared to the prior year quarter, partially offset by a $1.9 million restructuring charge in the second quarter 2018 for employee termination costs associated with the consolidation of the operations of our Peninsula Insurance company subsidiary into the home office. As we announced earlier, we expect to achieve annualized expense savings of approximately $3.7 million as a result of implementing the Peninsula consolidation.

In total, our second quarter combined ratio improved modestly to 105.6% from 106.4% in the prior year quarter. The restructuring charge added 1.0 percentage point to our current period combined ratio. Our underwriting profitability fell short of our target, but we believe all of the actions we've taken will result in improved performance in the second half of 2018 and into 2019.

Moving to investments. Net investment income increased 12.3% to $6.3 million for the second quarter 2018. The increase in net investment income primarily reflected an increase in average invested assets relative to the prior year second quarter. Net realized investment gains were $1.5 million in the second quarter 2018, compared to $1.1 million for the second quarter 2017. As a reminder, for 2018, new accounting guidance requires that we mark our equity securities to market value at each quarter end with the change in market value flowing through net realized investment gains.

Our total invested assets exceed $1 billion with approximately 90% invested in high-quality fixed maturity securities. We are staying the course in terms of an investment strategy with new money investments primarily within the categories of corporate bonds and mortgage-backed securities. In summary, we reported a net loss of 790,000 or $0.03 per Class A share for the second quarter 2018, which included the $1.3 million after-tax restructuring charge I mentioned earlier. That compared to a net loss of $2.3 million or $0.08 per Class A share for the second quarter of 2017. Our book value per share was $14.85 at June 30, 2018, compared to $15.95 at year-end 2017, with the decrease reflecting the net loss for the first half and unrealized losses in our available-for-sale bond portfolio due to an increase in market interest rates. Our Board of Directors declared quarterly cash dividends of $14.25 per Class A share and $12.5 per Class B share payable August 15 to shareholders of record on August 1.

With that, let me turn it back to Kevin for closing remarks.

Kevin Burke

Thanks, Jeff. We believe that our management team has made considerable progress on some key initiatives to improve our underwriting performance, and we expect more favorable results for the remainder of 2018 and, particularly, in 2019. Our core values include maintaining a conservative underwriting culture and pricing discipline, continuing our investments in technology and maintaining a conservative investment approach to deliver value to all of our stockholders. Our entire management team is focused on long-term strategy that adheres to this core philosophy.

We are focused on optimizing our business mix to take advantage of both of our diversification of products and our strong agency relationships. We continue to see favorable opportunities in our commercial segment, and we are aiming towards a shift to a higher proportion of commercial business. To achieve this objective, each of our regional Vice Presidents and subsidiary Presidents have prepared a plan that includes three year premium growth projections and action steps they will take to achieve the market, target market for the business. We will be further refining those plans as we formulate our business and strategic plans throughout the remainder of the year.

Another initiative is the continued enhancement of technology we utilize in all aspects of what we do. We intend to further enhance our predictive modeling capabilities as part of our ongoing systems modernization. Our IT management team has expended significant effort over the past several months to evaluate alternative approaches, vendors and systems, to develop a comprehensive plan to replace our remaining legacy systems. We are working with IT consultants to develop a comprehensive blueprint for systems that will meet our needs into the future. We are focused on underwriting profitability and book value appreciation over time, while preparing our company for the future.

With that, we'll ask the operator to open the lines for any questions that you may have.

Question-and-Answer Session

Operator

[Operator Instructions] Your first question comes from Christopher Campbell with KBW. Your line is open.

Christopher Campbell

I guess, my first question is on just personal lines' year-over-year net premium growth, which is kind of the slowest since we've seen in a while. So just presumably reducing the premium should take out more of the loss content, but how are you thinking about personal line expenses if this premium growth slows?

Jeff Miller

Well, we obviously are monitoring expenses very closely. The question really is what are we doing in terms of operations and what are we going to reduce those expenses as the premiums would be modestly lower. We are looking at things such as the reports that we use in the automated underwriting processes and the coding processes because as we slow the new business growth, we do expect to have a lower level of coding activity, which should result in lower expenses related to the ordering of those types of reports. We've also reduced the incentives that we were paying to our agents.

For personal lines, we at this point do not have any new business incentives in place for personal lines that will result in a reduction of our costs. We've implemented a loss ratio component in agency incentives that are related to production in total. So that will also have, we believe, an effect on our overall cost structure. But you raise a very valid point that we are continuing to analyze regional operations, looking for efficiencies wherever we can, the consolidation of the Peninsula operations in Salisbury into the home office was an example of a way that we will reduce the overall expenses for the entity. So those are the types of things that we are doing, but we continue to look for opportunities to reduce those costs.

Kevin Burke

Chris, as it relates to Peninsula, we had basically six underwriters, that is, now that work is being done by two underwriters as it's been centralized here in the home office. We think that there are some efficiencies that have been gained, but obviously, from a cost standpoint, we also think that it's very efficient. We recognize the fact that we're going to have to lower that expense load, particularly as it relates to personal lines.

Christopher Campbell

Got it. And so how would you be, and, Kevin, I think you had mentioned kind of your second part of the script was that you're thinking about pivoting more towards commercial lines, and I know Donegal historically has tried to target like a 50-50 ratio. So what are you thinking in terms of how that target could look going forward, commercial versus personal lines? And would you be able to, I guess, transfer some of the investments that you've made on the personal line side and kind of repurpose them for the commercial side?

Kevin Burke

Our goal, Chris, is really to achieve a 60-40 split, where 60% of our total book of business is commercial. In order to do that, it's somewhat sort of like turning an aircraft carrier. You can't do it dramatically. You can't do it within 18 months. You can't severe or create disruption within your agency base. And so what we want to do is we put a plan together that in a very methodical way over the next 2.5 to 3 years, we'd like to be at that 60-40 split. We have actually started a repurposing process, if you will. There are some very good solid underwriters in perhaps some of our personal lines areas that we know that we need to augment the commercial back office with.

And so we do see that as an opportunity to take some very skilled Donegal underwriters and sort of shift them over into the commercial line side to support that need and growth in the future. And we have been very active for the last 7, 8 years, bringing in a trainee group every year for the commercial line side of the house as we continue to build bench strength, and we recognized years ago that, that would be a problem had we not taken that initiative. So that is yielding some very good results as we have sort of changed that demographic of our commercial lines underwriting staff. We have a very strong bench now. We have additional trainees that are in place that are learning the business, and we do have resources on the personal lines side that could always transfer over as that commercial lines book of business continues to grow.

Christopher Campbell

And so I know that you guys have done, Donegal Mutual, not the publicly traded group, had done like an association with Mountain States. So how much would that factor into as that enters the pool in the upcoming years presumably? How much would that factor into the commercial and personal line split? And would you also be looking at maybe inorganic opportunities, possibly with the proceeds of the bank sale. I mean, could that be a way that you could accelerate that, that shift that you're looking for?

Jeff Miller

Certainly, the Mountain States affiliation, we expect that will be part of the overall plan to increase the commercial mix, the portion of our mix of business that is commercial business, because as you recall the Mountain States operations in the 4 Southwestern states of New Mexico, Texas, Colorado and Utah, that's 100% commercial lines. That transition, just while we're talking about, is going very well. Our systems are in place in all 4 states, or 3 of the 4. The fourth will be online here within the next few months. And the Donegalization, as we like to call it, of that operation is nearly complete.

We are now through the 1-year mark. And so the re-underwriting of that book of business is essentially complete, and we are seeing a nice flow of new business coming into that entity. As you said, it's a Donegal Mutual affiliation that Donegal Group currently does not participate in the premiums or the underwriting profits of that entity, but it's our intention to, at some point, include that business in the pool between Donegal Mutual and Atlantic States, which would then bring 80% of that revenue stream into Donegal Group Inc.'s revenue. So we do expect to see that coming in within the next year to 18 months.

We're continuing to evaluate the timing of that, but we are encouraged by the progress and that definitely would lead into the plan to shift the mix of business towards commercial lines. As to the bank proceeds, we will be investing those proceeds in the furtherment of our insurance operations to shore up the surplus at some of the subsidiaries, where growth has been especially strong to make sure that there is adequate surplus there from a statutory level to support the premium growth and that will be invested to support the commercial lines growth that we expect over the next two to three years.

Christopher Campbell

Got it. And would any of that be, would any of those proceeds be used to pay down any of the existing line of credit?

Jeff Miller

That's certainly under consideration. Yes, more than likely, we would utilize those proceeds to lessen our interest expense.

Christopher Campbell

Got it. And then just one last one, if I may, just on workers' comp. I know, I think the last three quarters you kind of had a year-over-year premium decline. So what are you seeing in that book? I know this quarter had a little bit of noise with the unusually high large amount of, like, severe losses. But what are you seeing in your attritional book just in terms of frequency, severity, exposure, growth and where are you seeing those [indiscernible] exposures coming in?

Kevin Burke

Let's take the first part of your question from a growth perspective, Chris. It really simply comes down to competition. We have a lot of monoline workers' comp writers out there that are really refining their rates and being very, very aggressive. We are not going to chase underpriced business. And worker's comp, as we all know, have been very, very profitable for us. It's been generally profitable for the industry for the last three or four years. And as carriers are dealing with loss ratio issues that they have in commercial auto, they are simply chasing the workers' comp piece for the profitability level. So we are having some reduction in overall policy count and growth in workers' comp, and it really comes down to simply the competitive nature of it in the rating. And we think that it's in our best interest to make sure that we sort of hold the line and not necessarily just chase accounts for lower rates because we all know where that gets you. So unfortunately, the industry continues to erode the profitability levels of workers’ comp. How long that's going to continue or where it bottoms out is really anyone's guess, but we are actively working with the agents and the accounts to do everything possible to retain the business, but the rates are getting very, very competitive right now.

Jeff Miller

Chris, this is Jeff. Just to follow up on Kevin's comments and address your question about the losses. We did see some, just a handful of larger losses in the current quarter. When you look at our experience over the last several years, it's typical for us to have one or two what we would characterize as catastrophic injury claims and generally, one or two in a year is kind of what we've seen. But we actually had four of those in the current quarter. Unusual circumstances surrounding each loss so that there's nothing that you could necessarily underwrite around, but they were just some of those what we call shock losses that hit the current quarter. You look at our results for the first six months, we had an 88.1% combined ratio, that's combined ratio in workers' comp. So it's still performing extremely well. 92.9% in the current quarter reflects that handful of losses. But as far as the underlying frequency, severity trends, other than those unusual claims, there is really no change. Everything is very stable.

Christopher Campbell

Got it. And these are, and these, and the four catastrophic claims that you had this quarter, that's not due to like new employees or the economy picking up, it's not like that type of trend or anything?

Jeff Miller

No, not at all. I mean, 1 was an auto accident. There were other injuries that would be what you would typically see. The 1 employee was a long-term veteran employee of the company. So I can't really tie it to economic conditions or employment necessarily. They were just kind of those freak accidents that are very unfortunate, but occur from time to time.

Operator

[Operator Instructions] The next question comes from Bob Farnam with Boenning and Scattergood.

Bob Farnam

So I have a few questions on pricing versus loss cost trends. And so I'm going to assume that your, in your impression, pricing is exceeding loss cost trends. I just want to know by how much.

Jeff Miller

Sure. This is Jeff, again. I mean, the indications that we're getting from our actuaries on a state-by-state, product-by-product level are really driving our rate increases. And I would say in many of the states that we are able to file for rate increases that are close to or exceed our indications. That's not the case in every state. We do have several states where we are still playing catch-up and we have filed what we believe is the highest rate increase that we can get approved based on our experience with the regulators in those states and viewing what our competitors have done. But for the majority of the business, the larger states, the states where we have the highest concentrations of business, we're generally taking increases that are in line with the indications. But there, as I said, are exceptions to that. And we believe that within 1 more rate cycle, we should be where we need to be in terms of getting the rates to be where the actual indications are saying they should be.

Just to give you a sense of the rate increase, component of our premium growth in the current quarter might be helpful. In personal lines, this is on a direct basis. Our rate increase, increase in premiums written included a 6.7% rate increase and that was offset by attrition that exceeded our new business and writings. And in commercial lines, it was 3.7% rate increase, again, which is higher than or very close to the 5.8% net premium written increase. But in commercial auto, the rate increase was 8.5%. So in the lines where we've had some difficulty in profitability, we are seeing a much higher rate increase impact in our net written premiums and, of course, that will eventually find its way into the earned premiums over the next six to nine months.

Bob Farnam

All right. So the personal lines rates up 6.7%. I'm assuming that's mostly auto, is that the case?

Jeff Miller

It's split between auto and homeowners, very close, both lines are very close to that number.

Bob Farnam

Okay. So you did note some adverse development in personal auto. You booked a 110 combined ratio there. So I suppose I can do the math afterwards, but just do you have an idea what your actual, your combined ratio would have been if we take that development out in personal auto combined ratio?

Jeff Miller

I'm sorry, I didn't run the math either, but it's about $1.8 million. So if you take that over, that's $64.8 million of premium, [indiscernible].

Bob Farnam

And the commercial auto combined ratio was 116. Now you didn't really say there was anything about development there. Is that 116 a good accident year number that we should be using or was there some development in there kind of behind the scenes?

Jeff Miller

There is actually favorable development in that line by about $1 million. So the current accident year loss ratio still is quite high and that reflects not only some large loss activity, continuation of the trends we've been talking about, but also a much more conservative view from our actuaries on future development in the current accident years. We are reporting a much higher accident year loss ratio than we would have, let's say, a year ago.

Bob Farnam

And with the auto rates going up 8.5%, some severity trend in there as well. That 116 is going to be elevated for a while until you start getting those rates really flowing through. So there could be another probably, more than one year probably for that to kind of come down into a more profitable range?

Jeff Miller

Right. We're not expecting significant improvement until middle part of 2019 at the earliest.

Bob Farnam

How much of the auto book is either personal or commercial, I guess, is stand-alone versus packaged?

Kevin Burke

There is not a lot of it that's stand-alone. In fact, initiative that we put in place at the beginning of 2018 is not really write any commercial monoline. Most of it, Bob, is package. And on the auto side for personal lines, there was incentives and have been incentive put in place in prior years to really package around those accounts. So we worked very, very hard to make sure that any monoline auto, commercial or personal lines, that we've stayed away from that, deemphasized it.

Jeff Miller

Yes, just to follow up on that. The only exception to that would be several states where we pulled out of the homeowners market a number of years ago and we continue to write auto only, and we're currently evaluating those states and really looking hard at the profitability levels there.

Bob Farnam

All right. Okay. And last question that I have is, so you're implementing predictive analytics. Sounds like you're getting into the, you've done the auto lines, you're getting into other lines, what type of impact do you expect or did you plan on having when you roll those products out on your loss ratio?

Kevin Burke

The first thing, Bob, is that we've really gained a lot of confidence in this model. It's been out for now couple of years. We're seeing the results of that model and how the underwriters use it and how we apply it. And so there is confidence there. And so with that, we start to roll it out in some of the other areas. One of the items that I mentioned initially on the call is the fact that one of our key initiatives going forward is really utilizing technology to the, its fullest extent, and we will be deploying some of the proceeds in the capital as a result of the sale of the bank into those sorts of initiatives. We will continue to roll that out. We have just recently augmented the private passenger auto model with some mileage component to it as well, that helps further refine that book of business. And so I think that what you will see is over the next several quarters, as we implement these models in different lines of business, it really helps us define that book, and I think, in particular, you will see it in the private passenger auto book. A year from now, we should be sitting here with, in a much better position because we would have been a full year of applying that model in addition to the rates. So we're optimistic about it. We've got it on our key objectives for 2019 to really further expand that, and we're looking forward to the results that it would yield.

Jeff Miller

Yes, Bob, just following up on that. To be a little more granular, the loss ratios are quite a bit higher, in the higher deciles. So when you look at the model scoring, we would say an 8, 9 and 10, those scores, the business that is scored in those higher deciles has a much higher loss ratio. And so to the extent that we're not writing new business in those deciles and to the extent that we can reduce the overall percentage of our business that is in those higher deciles will have a dramatic impact on the loss ratio. So much more dramatic than what the rate increases will have. So we do have a lot of confidence and expectation that the actions we're taking as a result of the predictive model scoring and being more aggressive in our usage of the scores will have a solid impact on the profitability going forward.

Robert Farnam

So on the personal auto side, I mean, I think if I do the quick math, maybe you had an accident year combined ratio of 107 or something like that. You're giving me rate increases you do in the predictive modeling. So is it fair to say that you're expecting that combined ratio to come back to even or better, to break even or better next year?

Jeff Miller

We will be very disappointed if we're not seeing combined ratios under 100 by the middle part of next year.

Kevin Burke

We've run some projections, Bob. The actuaries have sort of run it out over the next year to 18 months, and we are on track to sort of get to that breakeven point with that particular line of business.

Operator

We have no further questions at this time. I turn the call back to the presenters.

Jeff Miller

We want to thank everyone for joining the call today, and we look forward to speaking with you again in a few months after the release of our third quarter results. Thank you, everyone.

Kevin Burke

Thank you.

Operator

This concludes today's conference call. You may now disconnect.