Image source: The Motley Fool.
State Auto Financial Corp (STFC)
Q3Â 2019 Earnings Call
Oct 31, 2019, 11:00 a.m. ET
Contents:
- Prepared Remarks
- Questions and Answers
- Call Participants
Prepared Remarks:
Operator
Welcome and thank you for standing by. [Operator Instructions] . If you have any objections please disconnect at this time.
I would now like to turn the call over to State Auto Financial Corporations Director of Investor Relations Natalie Schoolcraft.
Natalie Schoolcraft -- Director of Investor Relations
Thank you Latania. Good morning and Happy Halloween, everyone. Welcome to our third quarter 2019 earnings conference call. Today I'm joined by our Chairman President and CEO, Mike LaRocco; Senior Vice President and CFO, Steve English; Senior Vice President of Personal Lines, Jason Berkey; Senior Vice President of Commercial Lines and Managing Director of State Auto Lab Kim Garland; Chief Actuarial Officer Matt Rubik; and Chief Investment Officer Scott Jones. After our prepared remarks we will open the lines for questions. Our comments today may, include forward-looking statements which by their nature involve a number of risk factors and uncertainties which may affect future financial performance. Such risk cars may cause actual results to differ materially from those contained in our projections or forward-looking statements. These types of taxes are discussed at the end of our press release as well as in our annual and quarterly filings with the Securities and Exchange Commission. Financial schedules containing reconciliations of certain non-GAAP measures along with other supplemental financial information are included as part of our press release and available on our website stateauto.com under the Investors section.
Now I'll turn the call over to STFC's Chairman, President and CEO, Mike LaRocco.
Michael E. LaRocco -- Chairman, President and Chief Executive Officer
Thanks Nat and Happy Halloween everyone. The story of this quarter is validation expanding success and focus. Let's begin with the validation. When our transformation began four years ago the single biggest decision we made was to become a digital-only carrier. Within the industry this was unique and within the agency distribution system it was a risk. With agents accept a platform that did not allow for cash checks or paper one that required an electronic signature use of the checking account or a credit card. For the last three years, we have clearly proven that this new approach will work within personal insurance. After years of declining volume and poor results our Personal Lines book is growing again. The question of acceptance was again raised as we built a new platform both commercial auto and small commercial or BOP. It was one thing to think it would work across personal insurance but what about commercial? In our earliest days last year the answer to that question was not clear. However this quarter was the fourth consecutive quarter of record growth across these lines and there is no longer a question.
Our decision to become a digital-only carrier has been validated not only in personal insurance but commercial and auto and BOP as well. This was a significant achievement and another critical step as we rebuild state auto from a traditional regional insurance carrier into an innovative property and casualty carrier that is built for emerging innovation and long-term success. The second part of this quarter's story is our expanding success. Early on we made hard decisions to exit a number of lines of business large commercial trucking programs in E&S casualty and property. At a time when we were shrinking across personal and commercial this was another bold decision. It's always difficult, to make the call that results in a smaller company. I have watched as competitors have rates in the Specialty business large commercial and even reinsurance it's too soon to decide which approach will work. But for state auto we know who we are. And our goal is to be exceptional in personal lines as well as small and middle market commercial lines. This quarter was a key indicator that our focus while causing some initial shrinking of our book is now paying off. We continue to grow in personal insurance. And as I noted above we are also rapidly growing rapidly in BOP policies in force and commercial auto.
That is not the entire story. Our middle market commercial business continues to grow and we have not yet launched our new digital platform. The improvement in this line as a result of leadership and culture changes. We have built an environment that allows our talented underwriters to effectively and efficiently evaluate and choose risk in partnership with our agents. We will launch our new platform next year which we believe will impact not only growth but more critically our efficiency. This quarter our growth across commercial lines both small and middle market has allowed us to demonstrate expanding success. We are now a successful write both, personal and commercial. I'd also note, we continue to grow farm and ranch. More importantly we are seen as an important market for these lines. Before I close with this part of the story I must mention workers' compensation. We continue to shrink in this slot this year albeit at a slower pace. But I feel this is indicative of our strength and the fact that growth without profitability is unacceptable. This is a challenging business growing as easy growing profitably is not. We will do both.
At this point we are on pace to pass $2 billion in premium this year. And remember that's the gross. As we continue to add scale which will soon begin to show in our expense ratio expanding our success was a major part of our third quarter story. The final piece of this quarter's story is focus. This part of the story can be seen in both our expenses and especially across personal auto. Let's begin with auto. This was the first product we brought to market in 2016. While the platform that issue of the business worked very well we were faced with a portal that did not meet our needs. The technology weakness in that quarter impacted sales service and ultimately retention. So, we decided to build our own internal formal. In addition, we have been adjusting and improving our pricing models and will take rate as needed to be certainly returned this large line of profitability. This quarter we began a focused effort to clean up the remaining technology challenges and make the latest updates to our model. While this has slowed our new business growth in auto we felt it was critical to focus on execution so that our customers' experience is outstanding and we can improve our efficiency.
These issues are not felt, within Commercial lines, as we initially launched those products with our own portal. While I'd like to be growing auto more aggressively it's imperative we do so profitably efficiently and with the highest degree of service to both our agent customer and policyholder customer. If the result of that focus is short-term slowdown in growth in this one line, I'm very comfortable with that decision. Once these fixes are in place I estimate three to six months at the longest we will once again accelerate our growth. And by the way we fully expect at least modest growth as we adjust the product and finalize the technology issues. The final piece of our focus is on expenses. As we track our ongoing investment in technology we can now see the resulting improvement in our efficiency and ability to grow effectively.
There is still much work to be done as I noted above but this quarter was another major step forward in our focus on efficiency which is the last piece in our drive to build a truly unique and effective PC carrier. I could not be more pleased with our progress to date. There's much work to be done but we are well positioned for the coming quarters and years. As innovation continues to change our industry. The true long-term winners will have made the infrastructure investments that will allow for the implementation of new technology to price sell and service as our industry finally emerges and embraces change.
With that I'll turn the call over to Steve.
Steve English -- Senior Vice President and Chief Financial Officer
Thanks Mike, and good morning everyone. For the quarter STFC reported $0.25 net income per diluted share with operating income of $0.34 per diluted share. This compares to $0.76 net income and $0.44 operating income for the third quarter of 2018 on the same per share basis. For nine months STFC reported on a diluted per share basis net income of $1.25 and operating income of $0.34. This compares to $0.86 net income per diluted share and $0.53 operating income per diluted share for the first nine months of 2018. One factor in the fluctuation of net income quarter-over-quarter and year-to-date over year-to-date is reported net investment loss or gain. Net investment loss or gain reflects the change in unrealized gains and losses on equity securities and other invested assets. For the quarter STFC reported a net income of $4.4 million in pre-tax net unrealized losses on investments. While in the third quarter of 2018 STFC reported $16.8 million of net unrealized gains. For the nine months ended September 30 2019 STFC reported $51.2 million of net unrealized gains on investments compared to $11.1 million of gains for the same period a year ago. As a reminder non-GAAP operating results exclude net of tax net investment gain or loss. The quarterly GAAP combined ratio of 99.5% was higher compared to 98.4% from the third quarter a year ago. The cat loss ratio was down. The non-cat loss and ALAE ratio was up while the expense ratio improved.
For the first nine months of 2019 the GAAP combined ratio of 103.6% compared to 102.6% for the same period in 2018 reflecting 1.3 points higher cat losses and non-cat loss and ALAE ratio up 0.5 point and an expense ratio improvement of 0.8 point. The GAAP expense ratio for the third quarter of 2019 was 34.3% and for the first nine months 35% compared to 36.1% and 35.8% respectively a year ago. During 2019 we continue to build and roll out commercial lines technology and products. Having said that the quarter-over-quarter improvement was more than half driven by reduced consulting spend primarily it related and revised estimates of associate variable compensation with the balance of improvement across several other categories. Similar factors drove the year-to-date improvement with more coming from reduced estimates of variable compensation. On a statutory basis personal and commercial reported in the quarter a combined ratio of 98.7% compared to 96.4% for the third quarter of 2018. On a year-to-date statutory basis personal and commercial reported a combined ratio of 102.7% compared to 100.8% for the same period a year ago. Catastrophe losses during the third quarter were lower than a year ago while on a year-to-date basis the cat ratio for the first nine months of 2019 is 7.8% compared to 7.6% for the first nine months of 2018. Our estimates of prior year reserves continue to develop favorably overall but at lower amounts which was not unexpected.
For personal and commercial lines in the quarter $15.2 million was reported relating to non-catastrophe losses in ALAE compared to $19.9 million in the third quarter of 2018. For the nine months ended September 30 2019 favorable development totaled to $51.7 million compared to $56.4 million for the first nine months of 2018. Personal and Commercial's current accident year non-cat loss and ALAE ratio in the quarter was up 4.2 points 61% in the third quarter of 2019 compared to 56.8% for the third quarter of '18. And through nine months the current accident year non-cat loss ALAE ratio increased 0.9 point 59.9% for the first nine months of '19 compared to 59% for the first nine months of 2018. As we have discussed previously reserve estimates can be volatile from quarter-to-quarter based on many factors. And additionally the second third and fourth quarter third quarter of any given year can be impacted by the reassessment of that year's accident year pick relative to book loss ratios from earlier quarters. Jason and Kim will get into more specific product detail in their prepared remarks.
The statutory expense ratio for commercial personal and commercial improved 1.7 points as compared to the third quarter of 2018 and 0.6 point on a year-to-date basis. The same factors I mentioned earlier regards to the GAAP expense ratio drove the improvements. Items to point out regarding specialty runoff. For the quarter net written and earned premiums reflect some reinstatement premium $300000 related to an increase on a prior year loss that we ceded to our reinsurance partners. $0.9 million of net favorable development of prior year non-cat loss reserves in the quarter which now stands at $2.4 million year-to-date. And the impact of specialty on the overall statutory expense ratio has fallen to 10th of a point for the first nine months of 2019 as compared to 1.3 points for the first nine months of 2018. Net investment income was lower for the quarter due to lower new money yields the impact of lower rates on our mortgage-backed securities a slightly smaller fixed income portfolio and the refinancing of the notes with our current company which took place in the second quarter of this year.
And with that I'll turn the call over, to Jason.
Jason Berkey -- Senior Vice President of Personal Lines
Thanks Steve and good morning everyone. Personal Lines finished the third quarter with a combined ratio of 100.6% compared to 92% for the same quarter last year. The personal lines loss and LAE ratio was 71% in the quarter compared to 61.1% for the third quarter of 2018. The increase in the loss ratio is due to lower levels of favorable development in personal auto and homeowners as well as the impact of a shorter average tenure in the order book and higher non-cat weather and large losses in homeowners. Net written premiums for the Personal Lines is up 7.4% versus third quarter '18 reflecting higher levels of homeowners and other personal lines new business offset by a decline in personal auto premium. At the same time our expense ratio for the quarter was 29.6% compared to 30.9% in third quarter '18. Looking specifically at personal auto the statutory auto loss and LAE ratio in the quarter was 73.9% with a statutory combined ratio of 102.8% compared to 60.1% and 89.6% respectively in third quarter '18. The personal auto loss and LAE ratio results were impacted by a 5.9 point decrease in favorable development across multiple coverages including bodily injury along with some adverse development of UM/UIM. The earned premium from our connect product now exceeds the earned premium from our legacy product and we expect to continue to see pressure on our overall loss ratio from a shorter average policyholder tenure until our retention increases.
The BI frequency trend, on our legacy connect and connect books continues to be favorable. We are however seeing an increase severity trend in our connect auto physical damage claims driven by higher repair costs on newer model years. At the same time we have experienced an increase in the Rum and UIM losses in the 2019 accident year due to higher frequency in our connect book. Our recent personal auto rate actions have been more targeted than our more aggressive rate actions in 2018. We reflecting the moderation of BI loss trends, and we have begun to see increases in both our connect retention and our legacy auto retentions as a result. Overall auto retention was roughly 66% in third quarter of '19. We expect to see our auto retention increase as we have greater renewal rate stability with fewer rate changes and with less than 1500 legacy Georgia auto policies left to non-renew. In the third quarter '19 net written premium for personal auto was down 3.3% versus third quarter '18. Policies in force finished 6% below the third quarter '18 level and new business accounts for the quarter were down 18.6% over third quarter '18. To restore profitable auto growth we are rolling out advancements in our auto rate segmentation to improve our rate competitiveness as well as rolling out ongoing system enhancements to improve the ease of use of our connect product. Moving on to our homeowners product results.
The homeowners loss, and ALAE ratio was 5.5 points higher this quarter in the same quarter a year ago. The increase was driven by higher levels of non-cat weather and large loss activity. Our homeowners third quarter '19 loss and LAE ratio was 69.6% with a combined ratio of 99.9% compared to 64.2% and 96.8% respectively in third quarter '18. The homeowners cat loss and ALAE ratio for the third quarter '19 was 9.2 points which is 8.2 points lower than the third quarter '18 homeowners cat loss and ALAE ratio of 17.4%. The third quarter '19 non-cat loss and ALAE ratio of 54.6% was 13.7 points higher than the 40.9% in third quarter '18. Again driven by higher non-cat weather and large losses with less favorable development in third quarter '19. We continue to see opportunity for profitable growth with our Connect homeowners digital product, and anticipate the enhancements being made on our auto product will benefit homeowners as well. In third quarter '19 our homeowners policies in force increased by 12.1% over third quarter '18. New business counts in the quarter were up 16.8% over third quarter '18 and homeowners net written premium increased 19.5% versus third quarter '18. In terms of homeowners retention the quarter ended with home retention at roughly 75%. And we have made changes to our connect rate structure to improve renewal rate stability and we expect to see improvements in our retention as those changes renew into our connect homeowners book of business.
In conclusion, our Personal lines storyline for the quarter can be summarized as: first more targeted auto rate activity relative to the prior aggressive rate actions we have taken in recent years along with ongoing system enhancements are being implemented to increase our close ratio and drive new business growth. Second the impact of a shorter average tenure in our auto book resulted in higher loss ratios in the quarter versus third quarter '18. Ongoing efforts to increase our retention as well as underwriting actions and claims operational actions are targeting improvement in the auto loss ratio. And last homeowners combined ratio for the quarter was impacted by higher non-cat weather and large losses. We have plenty of room to continue to improve and we are focusing on initiatives that drive better core business results with confidence in what we can achieve in the fourth quarter and beyond as we focus on the fundamentals.
With that I'll turn the call over to Kim.
Kim Garland -- Senior Vice President of Commercial Lines and Managing Director of State Auto Labs
Thanks Jason. The commercial results are as follows: third quarter '19 combined ratio of 95.9% versus 103.7% in 3Q '18 and a 3Q '19 written premium increase of 16.8% versus 3Q '18. For the commercial business as a whole the story is: the statutory non-cat loss and ALAE ratio was good at 49.6% and was 3.9 points lower in 3Q '19 versus 3Q '18 with the 3Q '19 non-cat loss and ALAE ratio being lower in all commercial product lines except commercial auto which was 4.2 points higher than Q3 '18. Commercial loss ratios continue to be where we generally need them to be and commercial expense ratios are not and are still our biggest inhibitor to overall profitability. But we are starting to see progress on the commercial expense ratio. The 3Q '19 statutory commercial expense ratio of 38.9% was 2.9 points lower than 3Q '18 due to both lower variable compensation for associates in the quarter and signs of positive impact on the expense ratio from our efforts to build a more efficient commercial organization. Growth continues to strengthen for our Commercial Lines business. 3Q '19 was the fourth record new business quarter in a row for Commercial Connect that is commercial auto and BOP.
Commercial auto, new business premium is up 261% 3Q '19 versus 3Q '18 and total commercial auto premium is up 45% versus Q3 '18. BOP new business counts were up 92% 3Q '19 compared to Q3 '18. BOP new business premium was up 57% 3Q '19 versus 3Q '18. Average BOP premiums in connect are around 60% of what they were at pre-connect and total BOP premium was up 3% versus 3Q '18. Middle Market Commercial 3Q '19 new business written premium was up 90% versus 3Q '18. This is our seventh consecutive quarter of middle market new business growth of 24% or higher and middle market total written premium was up 30% versus 3Q '18. Farm & Ranch had a solid quarter of written premium growth at 6.5%. Workers' comp growth has started to stabilize with 3Q '19 written premium down only 1% versus Q3 '18. Here's an update on the path to a significantly improved expense ratio for commercial lines. Here's our current progress. In 3Q '19 the commercial expense ratio of 38.9% is 2.9 points better than our 3Q '18 commercial expense ratio of 41.8%. And our 3Q '19 year-to-date commercial expense ratio of 39.9% is 1.2 points better than our 3Q '18 year-to-date commercial expense ratio of 41.1%. We have previously discussed that to achieve a significantly better expense ratio in commercial Lines will require two things: improved unit economics across the different tasks that are done within our commercial business and more scale. Improved unit economics. Connect continues to give us better unit economics pre-connect commercial auto and BOP had a 0% straight-through processing percentage from new business as we manually touched every piece of new business. In a connect world these products are achieving around a 70% straight-through processing percentage.
With connect, and other efficiency gains the commercial division is handling 17% more total direct written premium in 3Q '19 versus 3Q '18 and 104% more direct new business written premium in 3Q '19 versus 3Q '18 with fewer commercial associates than a year ago. We need to get more of the commercial premium on this more efficient platform. For all of commercial lines a percentage of premium on connect has been the following: for all of 2018 it was 1.8%. For 1Q '19 it was 5.2%. For 2Q '19 it was 6% and for Q3 '19 it is 6.4%. 2020 is when this metric should really start to accelerate with the rollout of Farm & Ranch Connect and Middle Market Connect CPP in early 2020. Scale as previously mentioned we are starting to see the scale issue be solved. Commercial auto at 45% written premium growth rate for 3Q '19 versus Q3 '18 with connect being the main driver. Middle market a 30% written premium growth rate. This is from a second order impact from small commercial Connect as it freed up time for commercial underwriters to focus on middle-market risks. The impact of the Middle market Connect launch in early 2020 is still to come. Farm and Ranch the 6.5% written premium growth rate. The first Farm & Ranch Connect state is scheduled to launch in 1Q '20 and connect will allow Farm & Ranch to enter eight to nine new states in 2020. Small commercial a 3% written premium growth rate at 92% new business count growth rate. Connect provides us the platform to grow but we have to continue to both increase unit counts and become more effective with larger BOPs to really increase this growth rate.
Workers' compensation a minus 1% written premium growth rate significant growth will likely not occur until connect launches in 2020. And again the key to this plan will be to ensure that loss ratios do not deteriorate as we work on these items. The commercial business results by product line are as follows. I'll focus on the loss and LAE results for each product line as the acquisition and operating expense ratios for every product line are poor as previously noted. Commercial auto. The commercial auto loss and LAE ratio in 3Q '19 is 66.4% which is an 8.1 point deterioration versus 3Q '18. The drivers of this deterioration are the following: 2.3 points come from a higher cat loss ratio this quarter versus a year ago 1.6 points from a higher ULAE ratio this quarter versus Q3 '18. We have five points of additional rate need on our legacy commercial auto book that we have not achieved. This is worth about four points on the overall commercial auto book as 80% of our commercial auto book is still legacy. And we have identified a few underpriced segments in our commercial Auto Connect version 1.1 pricing model. Smaller businesses and newer businesses are a couple of examples of these segments. Version 2.0 of the commercial auto pricing model was implemented last week in the first 13 states which should address these segments. The profitability of these segments or lack of profitability of these segments are worth about one point on the total commercial auto loss ratio.
Commercial aut,o written premium again, in 3Q '19 was up 45% versus 3Q '18. Small commercial package the small commercial package statutory loss and LAE ratio in 3Q '19 is 54.3% which is a 13 point improvement versus 3Q '18. Small commercial package written premium in 3Q '19 was up 3% versus 3Q '18. Middle market commercial. The middle market statutory loss and LAE ratio in 3Q '19 is 60% which is 3.3 points lower than 3Q '18. As with commercial auto we are keeping an extra close watch on middle-market loss ratio since the growth rates are so high. Middle market written premium in 3Q '19 was up 30% versus 3Q '18. Workers' compensation. The workers' compensation statutory loss and LAE ratio in 3Q '19 is 52.4% which is 11.2 points lower than 3Q '18. Our workers' compensation premium decline discussed in previous quarters have started to stabilize. Our workers' compensation growth rates by quarter this year have been the following: 1Q '19 versus 1Q '18 down 20%; 2Q '19 versus Q2 '18 down 12.5%; 3Q '19 versus 3Q '18 down 1%. The stabilization of our workers' compensation growth rate is a tale of two stories. For renewal written premium our workers' comp renewal written premium is still down significantly down 14.8% in 3Q '19 versus 3Q '18. We walked away from some large debit Mod renewals in 3Q '19 as we did not believe we could hit our profit margins at the prices at which competitors were willing to write these risks. As previously discussed we continue to be committed to maintaining underwriting and rate disciplined during this part of the cycle. New business written premium our workers' comp new business written premium increased for the second consecutive quarter in 3Q '19 and it was up 57.4% in 3Q '19 versus Q3 '18.
This new business premium increase is, being driven by continuing to improve how we integrate workers' compensation as part of a total commercial package sale and this improvement combined with the general increase in our overall commercial new business volumes is driving the increase in workers' compensation new business. We will likely not see significant growth in workers' compensation until; worker's comp connect is launched in the second half of 2020 but stabilizing the workers' comp premium volume is an important accomplishment by the team. Farm and Ranch. The Farm & Ranch statutory loss and LAE ratio in 3Q '19 is 50.7% which is 8.3 points lower than the 59% in 3Q '18. Farm & Ranch written premium in 3Q '19 was up 6.5% versus 3Q '18. We had been planning for the launch of Farm & Ranch Connect in 4Q '19 but are now expecting this launch to happen in 1Q '20. 3Q '19 was another important step in building a better future for commercial lines at state auto. We set connect new business records in the quarter for the fourth straight quarter. The new business success occurred in both small commercial Connect and middle market our middle market business and we saw another quarter of early impact on the expense ratio from our work over the last couple of years. A lot of sweat equity from the last couple of years went into producing this quarter's results of 95.9% combined ratio with 16.8% growth. I'm thrilled that the commercial team is starting to see the fruits of their labor and I continue to be incredibly proud of the work of our commercial lines team. But the team wanted me to let you know that they believe our best days are still ahead of us.
With that we will open the line for questions.
Questions and Answers:
Operator
[Operator Instructions] And your first question comes from the line of Larry Greenberg.
Larry Greenberg -- Janney Montgomery Scott -- Analyst
Good morning and thank you. So a lot of detail was given. But it appears that both auto lines are experiencing deteriorating profitability from an underlying standpoint this year. And I know in personal the lower tenure is having an impact. But just generally is that an appropriate perception that you have lost profitability in both auto lines? And maybe just a little bit more color on how we should be thinking about the next 12 months?
Jason Berkey -- Senior Vice President of Personal Lines
So I'll start with commercial lines. And I think there is some truth to your observation but some of it is probably our own fault too. I think on the legacy book Larry we have taken we are on track to take about five to six points of rate this year when we really need to take around 10-ish. And so we have sort of reinforce the message with our underwriters that they need to make sure that they get adequate rate in commercial auto. And so that deterioration is kind of on us of sort of keeping up with the rate need that we need to accomplish. And I think that is probably the biggest part and you should be aware that we sort of we are aware of the combination of trends in the marketplace.
And when you're growing commercial auto at the rate we are sort of managing whatever sort of reduction in average 10-year new business penalty you might want to have. The second piece is while connect is a much smaller piece of our book I don't think or to me it's not unexpected to at least found a couple of segments that we needed to dial in with the next set of rate changes. So as insurance people it's hard to overstate how excited we were to launch the next version of the model in 13 states last week. Pretty confident that those will address that and we are also making sure we keep up with the overall rate need on the connect side of the house.
Steve English -- Senior Vice President and Chief Financial Officer
And Larry on the Personal Lines side we are certainly seeing some profitability challenges on the Connect book when we look at our legacy book it's a very stable profitable book. I don't really see any rate need there. At this point and the retention has actually started to come back some of the aggressive rate actions are quite a bit behind us now. But on connect we anticipate our retention continuing to increase. It needs to get up in the mid- to low 80s to really reduce that pressure that we are seeing from the tenure. But we are not just waiting for that. We're taking underwriting actions including some system enhancements that we will get in by the end of this year first quarter at the latest that really improve kind of our underwriting activity execution. And there's also ongoing agency management portfolio management actions rather focus on a few areas that we have seen some loss ratio challenges.
Michael E. LaRocco -- Chairman, President and Chief Executive Officer
Yes. Just Larry this is Mike. A couple of quick comments from me on personal mainly. And I've talked about this before it's absolutely not excuse everything's on us. But we build a brand-new product. And so as you do that and you launch a new product again it didn't look anything like the old ones. You go through a series of challenges. It was exacerbated as I mentioned in my commentary about some of the AutoCanada a vendor that caused additional issues. But as we work through that the ongoing fixing of the model and getting the peers right. We did all this in the face of of course at that time a fairly high level of BI severity. So again I it's always up to you guys to interpret all this stuff.
But I mean first of all auto probably the easiest line to fix. And what I mean by that is that this is high frequency low-severity lines and it's all the vast majority of it is not all of it is model-driven and so I believe that this is just a process of enter retention of the legacy business better retention the connect business the connect new business is actually performing reasonably well, as we would expect but the mix of business is putting some pressure on the loss ratio. So I actually think directionally we are doing the right things. We've got to act maybe with a little bit more urgency. But I feel good about kind of where we are at right now. But I do appreciate the question.
Larry Greenberg -- Janney Montgomery Scott -- Analyst
Thank you.
Operator
Your next question comes from the line of Freddie Slicer with KBW.
Freddie Slicer -- KBW -- Analyst
Hi, good morning. I just wanted to start on commercial auto with the core loss ratio. I was just wondering if there are any specific states that are underperforming? And what he's currently seeing in terms of frequency and severity trends and sort of what is your loss cost trend assumption right now?
Michael E. LaRocco -- Chairman, President and Chief Executive Officer
So no particular states sort of stand out I would say probably regionally the what we call our southern region probably needs more rate than our other regions. So we are working with the underwriters there to make sure that that happens. I think from a loss cost trend perspective or sort of the combined frequency severity trends. I think we are pricing in around 10%. So we think in general we need to do that. And that is pretty I would say again Natalie may have to clean this up and give more specifics. But I think it's pretty evenly split between frequency and severity.
Freddie Slicer -- KBW -- Analyst
Okay. And then I'm not sure if I missed it but what rate increases did you take in the quarter? And then also on the I think you mentioned 5% of additional rate needed and now you're seeing some underpriced business on the connect platform how much of your total commercial auto premiums business enterprise business make up currently?
Kim Garland -- Senior Vice President of Commercial Lines and Managing Director of State Auto Labs
So I think on we took between five and six rates points of rate this quarter and we needed another five on top of that. So we probably needed between 10 and 11 is what we were aiming for. So that was sort of the difference there. On connect I think on the underpriced segments were I think newer businesses...
Steve English -- Senior Vice President and Chief Financial Officer
Well the vast majority of our book is on connect.
Kim Garland -- Senior Vice President of Commercial Lines and Managing Director of State Auto Labs
Yes. So legacy is 80% of our book connect is 20% of our book and the connect rate need or the impact of that was about one point on the loss ratio.
Freddie Slicer -- KBW -- Analyst
Okay. And then on the just sticking with commercial auto premiums were up 45% year-over-year. So how much of that was rate versus exposure growth? And should we be thinking at this of this level of growth is sort of a run rate over the next few quarters despite the core loss ratio deterioration?
Kim Garland -- Senior Vice President of Commercial Lines and Managing Director of State Auto Labs
Yes. So most of it was is unit growth. I mean we have been taking if it's 45% you probably have high single digits of rate and the rest of that is unit growth. We have I believe the plans to get commercial auto the growth trajectory we are on. We are comfortable with. And so I think that will sort of continue to go on for a little bit.
Freddie Slicer -- KBW -- Analyst
Okay. And just sorry go ahead.
Kim Garland -- Senior Vice President of Commercial Lines and Managing Director of State Auto Labs
Yes. I think if you think about most of our activity around commercial auto. I think you'll see most of those on the rate side. So one for underwriters on the larger risks just reinforcing that we need to get the rate that we need to get. And then as I said I think we rolled out the first 13 states of the new model by the end of the year we should roll out most of the other states there will be a couple of stragglers based on state approvals.
Steve English -- Senior Vice President and Chief Financial Officer
And remember when you talk about the growth rate and are we comfortable with it regarding the rate need? Most of that rate need was on the legacy book the new business is coming on connect. And as Kim mentioned our latest model on Connect was just launched in 13 states we will roll it out to the rest. So the new business coming on to connect is much more model-driven. And we have a lot of confidence in not only the rate there but the rate per risk-based on the modeling that's been built
Kim Garland -- Senior Vice President of Commercial Lines and Managing Director of State Auto Labs
And as Mike always daringly reminds me 45% on a small base is not necessarily that much.
Freddie Slicer -- KBW -- Analyst
Great, Okay. And then just on the commercial lines expense ratio it seems like most of the products are making good progress. But how should we be thinking about modeling commercial expense ratios in 2020 and '21? In which segments do you expect which products do you expect to drive most of the further improvement?
Kim Garland -- Senior Vice President of Commercial Lines and Managing Director of State Auto Labs
So, I think as, you think about which products I think you can think about them in two ways. So for commercial auto and small commercial their connect sort of the build and rollout is done. And so the sort of IT expense ratio for those product lines will burn off earlier than they do for farm and ranch workers' comp in middle market. So I would expect to see sort of more improvement in those product lines earlier than the others. As I talk about we will still in 2020 you'll still see we still have to build and roll out build and/or roll it out the Farm & Ranch workers' comp CPP so those expense ratios. We'll still have to absorb that in 2020. And then part of our focus as we try and sort of either get more business on this more efficient platform. Improve our processes is sort of keeping our headcount and fixed expenses flat as we put more volume on it. So our expectation is that gives us a bit of a tailwind.
Freddie Slicer -- KBW -- Analyst
Okay. Great. And then just switching to personal lines. Jason I think I had you mentioned targeted rate increases in personal order budgets. Wondering if you could put a number on the rate increases that you're taking? And how much of your premiums is it impacting?
Jason Berkey -- Senior Vice President of Personal Lines
Yes. The targeted rate changes as I mentioned will really be on the Connect product and our legacy personal auto business is performing quite well. It's stable. And we have addressed over the last couple of years aggressively since some of the trend issues but we are not seeing that anymore. So on connect obviously it's going to vary significantly by state but most states we are talking overall rate changes low single digits mid-single digits. But the targeting of the rate changes would be increased segmentation really looking to make sure that in sales where we are having loss ratio issues that we are putting more there. So there would be areas of the book that might get significantly more rate increase than that.
Freddie Slicer -- KBW -- Analyst
Right. And then so how should we be thinking about the personal core loss ratio going into 2020 given some of these targeted rate increases and the recent deterioration?
Jason Berkey -- Senior Vice President of Personal Lines
Well I think the pressure that we faced from the tenure will continue as long as we have this level of retention. But we continue to see that improve month-over-month both on legacy and on Connect. And the targeted rate changes that we have lined up for first quarter we will begin to shift the business mix the target rate changes we took in the first half of this year did that as well really reduce the amount of no prior business and some other business that was shorter tenure higher loss ratio and we anticipate that will also occur next year but I would see that we are going to face pressure in the auto line from a number what you need to look at is you have to look at the net loss ratio on personal auto and the legacy. And in both of those lines in that auto line that we expect to connect loss ratio to get better and we expect the legacy loss ratio to get better based on the actions that we are taking. And of course on the legacy side the seasoning and the increased retention.
So then it's kind of depending on how you model that out you've got to kind of fix for that kind of determine that mix between the connect business and the legacy business and you've got to kind of work. What Jason says there's going to be pressure. It's just simply because of the both loss ratios will get better. We'll have more business on the Connect side than we will on the legacy or think of it this way. More new business versus seasoned business and that's going to put a little bit of pressure on that process. Having said that we think long-term in this and we believe as these things start to work out over time when we increase the retention on both of these lines that we are going to be able to get it to where we need the loss ratio to be on personal auto. Very confident about that.
Freddie Slicer -- KBW -- Analyst
Okay. And then just lastly on investment income it was down quite a bit year-over-year and quarter-over-quarter. Just wondering if you're making any changes to the structural duration of the portfolio like the low interest rate environment?
Scott Jones -- Vice President and Chief Investment Officer
Freddie this is Scott Jones. Not really making any structural changes to the portfolio. I think what you've seen happen is just the result of a lower rate environment that we are in now and the lower new money yield that we are receiving on new investments and how that's impacted our existing holdings of mortgage-backed securities. So no real structural changes to the portfolio in terms of duration or anything like that.
Freddie Slicer -- KBW -- Analyst
All right, great. Thank you for the answers.
Scott Jones -- Vice President and Chief Investment Officer
Thank you, Freddie.
Operator
Your next question comes from the line of Paul Newsome of Sandler O'Neill.
Paul Newsome -- Sandler O'Neill -- Analyst
Great. Good morning, guys. First I want to thank my KBW colleague for asking the first 500 of my questions and someone needs to learn the lessons. But the only question I have is more of an accounting issue with the tenure. And I understand on the personal lines auto business. I understand there's a new business penalty that's usually a couple of points. But I would have expected to see if there was a 10-year issue that come through more on the written premium side and then on the expense side as opposed to the loss ratio side. And is that underneath the other stuff that the tenures affecting the expense line? Or am I just not thinking the accounting correctly?
Steve English -- Senior Vice President and Chief Financial Officer
Well. Paul this is Steve. On in terms of tenure on the expense side I think the impact would be primarily in commissions in the first year versus renewal commission rate. So if you're year one to connect right when it was all new business first year commission rate is higher than the renewals. So as time marches on and you get more of a work of the percentage of the book in renewal at that lower commission you will see that commission rate and therefore your expense ratio come down from "seasoning." But in regards to other types of I mean that's the only one I can think of that has a seasoning effect on the expense side and of course on the loss ratio side it's a very similar thing although new business penalty on personal auto is more than two points in terms of that business. So I think that's a little optimistic on new business facility. But I'll let others react to that as well.
Kim Garland -- Senior Vice President of Commercial Lines and Managing Director of State Auto Labs
Yes. I think the new business penalty is a little bit higher. Again I want to be really clear. The reason as you look at the way the product was built and you look at our ongoing improvement in the modeling. Our expectation is that: number one we can reduce that new business penalty that we are going to be able to better match rate the risk from the job. Number two we are offering telematics. Telematics is now approximately 20% or so of our new business. We're getting more of that for the people that are actually taking it. Telematics we do not anticipate the same type of new business kind of on that business because it's a different type of risk and a different type of payment. So as we continue this is going to be very hard for you guys to model because number one as we continue to grow the Connect business and we will and we continue to improve the modeling around this. As some of the assumptions that you may have traditionally made about new business versus legacy any amount of the new business penalty is going to get a little bit trickier.
But number one as Steve said the new business penalty is typically closer to 10 to 15 points than it is two points. I mean I think that's really important for you to kind of think about that as you go through the modeling process. Secondly we have an expectation that our legacy retention is going to continue to it's been very positive in the last 12 months or so we have been seeing a consistent increase in our legacy retention obviously excluding Georgia we knowingly decided to not renew that business which is the right thing to do. And then third we also expect to see the retention of the new business improved because as we have tweaked the model and improved our segmentation we believe that our mix of business will come in at a type of risk that will increase our retention and some of the technology issues which created self-imposed through our mistakes some billing problems and some retention issues.
Those things are getting fixed as we speak and will be done over the next couple of months. So all of those things together are why when I talked to you about focus that we had in the third quarter that's continuing in the fourth quarter. We're bringing all those component pieces together around personal auto and this is why we are optimistic that the change that you will see in terms of both our growth next year and are more importantly our efficiency and profitability we believe we will be trending in the right direction fairly quickly.
Paul Newsome -- Sandler O'Neill -- Analyst
Great, thank you. Appreciate it.
Kim Garland -- Senior Vice President of Commercial Lines and Managing Director of State Auto Labs
You bet.
Operator
[Operator Instructions] There are no further questions. I'll turn the call back over to you Natalie for any closing remarks.
Natalie Schoolcraft -- Director of Investor Relations
Thanks everyone for your questions for participating in our conference call and for your continued interest in and support of State Auto Financial Corporation. We look forward to speaking with you again on our fourth quarter earnings call which is currently scheduled for Thursday February 20 2020. Thank you and have a wonderful day.
Operator
Thank you. That concludes today's third quarter 2019 earnings conference call. Thank you for participating. You may disconnect at this time.
Duration: 54 minutes
Call participants:
Natalie Schoolcraft -- Director of Investor Relations
Michael E. LaRocco -- Chairman, President and Chief Executive Officer
Steve English -- Senior Vice President and Chief Financial Officer
Jason Berkey -- Senior Vice President of Personal Lines
Kim Garland -- Senior Vice President of Commercial Lines and Managing Director of State Auto Labs
Scott Jones -- Vice President and Chief Investment Officer
Larry Greenberg -- Janney Montgomery Scott -- Analyst
Freddie Slicer -- KBW -- Analyst
Paul Newsome -- Sandler O'Neill -- Analyst
