Welcome to The Progressive Corporation's fourth quarter investor event. The company will not make any detailed comments related to quarterly results in addition to those provided in its annual report on Form 10-K and the letter to shareholders, which have been posted to the company's website, and we will use this event to respond to questions after a prepared presentation by the company. The event is available via a moderated conference call line and a live webcast with a brief delay. Webcast participants will be able to view the presentation slides live or download them from the webcast site. Participants on the phone can access the slides from the Events pages at investor.progressive.com. In the event we encounter any technical difficulty with the webcast transmission, webcast participants can connect through the conference call line. The dial-in information and passcode are available on the Events page at investor.progressive.com.
Acting as a moderator for the event will be Julia Hornack. At this time, I will turn the event over to Ms. Hornack.
Thank you, Jason. Good morning. Today, we will begin with a presentation about Progressive's home business. Our presentation will be followed by Q&A with our CEO, Tricia Griffith, and our CFO, John Sauerland. Our Chief Investment Officer, Jonathan Bauer, will also join us for Q&A by phone. This event is scheduled to last 90 minutes. As always, discussions in this event may include forward-looking statements. These statements are based on management's current expectations and are subject to many risks and uncertainties that could cause actual events and results to differ materially from those discussed during the event. Additional information concerning those risks and uncertainties is available on our 2019 annual report on Form 10-K, where you will find discussions of the risk factors affecting our businesses, safe harbor statements related to forward-looking statements, and other discussions of the challenges we face.
These documents can be found via the investors page of our website, investor.progressive.com. It is now my pleasure to introduce our CEO, Tricia Griffith.
Good morning, welcome to Progressive's fourth quarter webcast. Well, we wrapped up 2019 with another banner quarter and banner year. We're very excited to start 2020, and obviously, you've seen January's results. We continue to be really excited about our opportunities around growth and profitability. As Julia said, today is about property, we're very excited to tell our story, both where we have come from and where we're going. Before I get into that, I've got three quick items. The first one, we heard you. Many of you asked for us to give the loss ratios associated with catastrophes, and as you'll see in the January earnings, we started to do that for each line coverage.
Hopefully, that will be more transparent for all of you to understand the effect of our underlying loss ratios, in any given month with any given catastrophe, especially in the more volatile line coverages like property. Second, as you likely read in the Form 10-K, Progressive and the minority stakeholders of ARX decided to conclude the acquisition a year early. We're spending $242 million to acquire the remaining shares of ARX, and we will close that, if all things go well, which we expect that to happen in April of 2020 versus 2021. It was funny, I was talking to Dave Pratt right before this, and he said he had so many emails from people at Progressive Home saying, "Didn't this already happen?" This is really something where we've been dating forever.
People assumed we were married, let's just get it done early, and we're really excited to move forward and really execute on our plan. The third item, which I'm sad about, but happy for her and happy for Progressive is, Julia Hornack has decided to take her talents down to St. Pete with Progressive Home. All of you have gotten to know her very well over the years. She is really our investor relations guru, and she's fielded a lot of your questions over the years, and we're going to miss her. We're excited for her in her next stage of her career. She's been a controller, she's been a product manager, and now she's moving on to work on specialty products, some vendor management, and some process management. Again, we're sorry to lose her from here, but so excited for Progressive. Congratulations, Julia.
Thanks, Tricia. Thanks a lot. To all of our shareholders, portfolio managers, and analysts, it's really been a pleasure getting to know you and to do my best to represent Progressive and the facts that we provide in all of our public disclosures. It's been a pleasure serving our executive team. It's truly extraordinary and a wonderful partner. Our wonderful partners. I'm really excited to take my talents to St. Pete, hopefully. I'm sure I'll be back to Cleveland quite a bit to see my friends here, but I look forward to making friends with all of my extraordinary colleagues down in St. Petersburg, so thank you.
I imagine you will be very diligent of when you come to Cleveland versus when-
Yes
you stay in St. Pete.
Yes.
I'll visit you in the winter.
Exactly. Yep.
Great. Let's get started. We changed the vision statement last year, usually when you change a statement two words, it doesn't really make a difference. This really did make a difference. As we started to think about a new vision statement, we started to think about how many bundled customers we had, home fit right in there. As we stepped back and thought about being consumers' number one choice, we really thought about why don't we have the word agent in there? This actually came from Heather Day because our collective consumers go into agencies, we want the agents to think of us first in each of their agencies. In fact, 60% of our business, when you look at commercial lines and personal lines, come from the agency channel. I'm an agency customer myself.
I wrote in one of the quarterly letters last year about how important this channel has been for us, especially going forward. For us, this key word, this one change, is really significant for our growth. In fact, recently, I had the opportunity to do a fireside chat at The Big "I" with the CEO, Bob Rusbuldt. The Big "I" is a national alliance of about a quarter million businesses that sell auto insurance products or insurance products, I should say. I talked with his board, which have representatives for each state, really, I told them about how important they are to us, really wanted to solidify our relationship, know that we want to get inside the hearts and minds of all of our agents, invest in them, think about the future of our collective customers. That word is very important.
The home word made sense because we believe in a dual-channel strategy. One of our pillars is to have broad coverage. We want to be where, when, and how customers want to shop for home insurance as well as auto insurance and others. Whether it's in the agency channel through our Progressive Advantage Agency or HQX, we want to be there. The vision statement changed with two words, pretty significant. I shared this chart with you last quarter, really to talk about the return on equity, we looked at a five, 10, and 20-year timeframe with Progressive compared to the S&P 500 and the S&P P&C Index. We outpace in all of those years, both of those indices, we're really proud of that fact.
What I also said was that under leveraged capital, we would return to shareholders, either in stock repurchases or dividends, 2019 was no different. This was the first year of our new dividend policy, our board declared a variable dividend of $2.25 per common share, which equated to over $1.3 billion returning to shareholders. In fact, if you add in the four quarterly $0.10 per common share fixed dividends, it equated to over $1.5 billion we were returning to shareholders. Although we don't have comparisons, I'm pretty proud, very proud to say that our return on equity for 2019 was over 31%. That is phenomenal. I'll obviously update you once we have more of the comparisons, we're very proud of this.
The question I would ask if I were an analyst or an investor would be, given the recent struggle with profitability, are you still glad that you had the ARX transaction? A couple of words very easily. Yes, absolutely, unequivocally, all those things. I'm going to give you a couple data points, and then Heather Day and Dave Pratt will really talk about our plans, and we're really excited about that. First of all, over the inception to date period, we've grown 11% at a 96 combined ratio. We've made money and we've grown. Do we want to grow more and have wider margins? Absolutely. A key part of this acquisition was for us to have access to those preferred customers in the agency channel.
In 2019, we believe that $750 million of our auto premium in the agency channel we would not have had but for having the partner home product to have that bundled preferred customer. We equate that to nearly $5 billion in lifetime earned premium, well below our 96 combined ratio. Those are a few data points that say, absolutely, this was meant to be. On the more softer side, which I think sometimes is lost when you have an acquisition, is the culture and the people and being able to really run as fast as we can collectively together to gain market share. As you know, I give credit to John Auer, the former CEO of ASI, or ARX, I should say, and Glenn Renwick, the former CEO of Progressive, of really starting that relationship.
We had fits and starts along the way of buy, build, partner, nothing quite worked. That relationship started, we started with a 5% interest, and right away we saw that these companies were alike. In fact, I always smile when I go down to St. Pete because in their first year of business, they have a plaque on their wall that says, "We want to be the Progressive of homeowners insurance." To me, that just meant something. It was almost a foreshadowing. In addition, we knew that we wanted with this preferred customer base, to be able to lengthen our auto POEs, and that has worked. We have comparison products we both built, umbrella as an example, where we would say, ARX's products is superior to ours, we'll merge those together and have even a better product.
I think more importantly, as of January, Dave Pratt started reporting into Pat Callahan, we're really starting to share resources, R&D, pricing, to get that depth of segmentation in property like we have in auto. I do want to thank John Sauerland for getting us to the final acquisition because he really was working on that for these last couple of years, and we're so excited to be able to wrap that up in April. Again, the values, the people, and the ability to really search for the great customers that we want and be able to achieve our ultimate vision is really exciting to us. I'm going to end by just talking about something that's probably the softer side as well, but I think it's key to our culture. That is a formula that I talked about in my annual letter to shareholders.
The first part of the formula is question everything. I think when you have a high-tenured company like Progressive, especially at the senior levels, sometimes you can be just surgically focused in growing. My goodness, we've been doing that incredibly. Partly, you have to step back and say, "Why do we do this? Why do we invest in this?" Question things that you said we may never do. A great example is create a homeowners company. After we, for lack of a better word, failed to do that decades ago, we said we would never be in the homeowners business. Never say never because things around us change. We knew we wanted to evolve as a preferred company, and we needed to do that. This was perfect.
What I ask really everyone at Progressive to do, and my team does all the time, is we always question and have really incredible debates to make sure that we're always thinking as this environment changes so rapidly. The second part of the formula is being an always-grow mindset. I talk about this being sort of a double entendre. For me, it's everyone should care about their personal growth. I actually had an interview take place last week, and the woman said to me, "What do you do when you get bored?" I said, "If I get bored, shame on me," because there's so much out there to read, to listen to podcasts, to look at TED Talks, to be in outside events.
We really want all 40,000-plus Progressive people to continue to be curious and think differently, because a lot of the best ideas come from the grassroots effort. We think if we always grow, ensuring the company will grow, and that's that double entendre. Question everything, always growth mindset equals an enduring business. For us, this is going to be our legacy for all of our senior leaders to the people who come after us. That is our job to make sure we have an enduring business for decades and decades and decades to come. Speaking of an always growth mindset let's talk about property, both growth and profitability. We have two guests today both of whom I think you've met. Dave Pratt, who is our Property GM down in St. Pete has been here for quite a while.
His undergraduate degree from Duke is in electrical engineering, and he has a Harvard MBA. He came to Progressive in 1991 as the product manager for New York. I can't go over his whole resume, but he's ran marketing, product development, and many, many products along the way. We asked him to take over as the Property GM right after Hurricane Irma in 2017, and he's been down there since doing a fantastic job. He has a great story to tell. Before that, I'd like to introduce to you Heather Day. Heather has her bachelor's degree in economics from Miami University, her master's in international relations from USC, and if that wasn't enough for the always growth mindset, she got her MBA from Wharton. She also has an extensive resume. Started as a product manager, has worked in recreational lines.
She was our marketing leader for Snapshot, most recently, our preferred marketing leader rolling out our Platinum for agents, which she'll talk about a lot today. Most recently, a couple of years ago, she was part of that swap that we arranged, she is the Head of Agency and Sales Distribution. Heather, why don't you tell us what you've been working on?
Thanks, Tricia. I will echo Tricia and reinforce that Progressive continues to invest in the independent agent channel. We recognize the value that agents provide to consumers that are looking for both product depth and local professional advice. Our commitment to working with independent agents has been longstanding, but it lines up well with recent market research and trends. The graph that you see on the left is from J.D. Power and shows that over the last 5 years, independent agents have made steady gains in purchase satisfaction compared to captive agents, or exclusive agents as J.D. Power would call them. The satisfaction levels today are 20 points higher, and what they found when they were digging deeper was that it was the preference for flexible product offers that really drove the largest performance gap amongst these channels.
This area of flexible product offers is a place where independent agents thrive based on their ability to place clients among several different brands, compared to a captive agent that simply lacks that flexibility. This change in purchase satisfaction correlates to market trends. There was a parallel shift in market share beginning in 2015, 2016, a time when most major carriers were taking rate increases. Consumers started purchasing more in the independent agent and captive channels where price comparisons could be made most easily. Direct carriers made gain in personalized market share, independent agent carriers held steady while captive carriers gave up share. Progressive has steadily grown our own share in the independent agent channel, and that's accelerating in recent years as we have added the home to the lineup.
If we look back at our recent growth, we were clearly well-positioned heading into a hard market, our overall agency results were boosted by our product segmentation and pricing. Progressive was steady. We were profitable and competitive during a period where many other carriers were taking larger rate increases. This provided an excellent opportunity to make headway on those auto and home bundles. The compelling auto position made the overall bundle more competitive. The demonstrated commitment to property with the purchase of ARX in 2015 and the subsequent launch of our Platinum program really gave our agents a reason to believe that they could partner with Progressive as a preferred bundle carrier. Indeed, as a share of our new business, bundles have grown fivefold over this period.
Looking at our independent agent share across these two products, in private passenger auto, we have gained two market points each year in 2016, 2017, and 2018, which put us at a 20-point share in 2018. On the home side, Progressive has moved into the number 6 spot with 3.5% of the market in 2018. Despite this strong market position, we continue to see upside, and the home product is key to that. Expansion of our addressable markets was really core to the Destination Era strategy, and I want to reinforce how critical that expansion is for our upside in agency. The table you see breaks down auto and home market premiums across channels by our market segments. For a quick refresher on these market segments, you've got Sams, looking for auto insurance as needed, very price sensitive. They tend to be more non-standard.
Your Dianes come to us with prior insurance, more financially responsible, but not homeowners. Wrights, these are homeowners that purchase home or auto insurance from us, but they do not bundle. Then finally, we have the Robinsons. These are typically the most preferred customers. They bundle home and auto, and they tend to look to agents for ease and confidence. Not surprisingly, due to our legacy as a monoline auto carrier, we tend to over-index on the Sams, the Dianes, and the Wrights. Despite our steady growth in bundles, we still have just about 2% of the independent agent Robinson market. This is lots of room to grow in a $59 billion market. Disruption in the captive market is putting more of that $97 billion of captive bundles in play as well.
While there is still lots of upside and work ahead of us, we are pleased with our results to date. We have continued our momentum in writing bundles with agents, with 27% growth in 2019. Part of that growth comes from the expansion of our Platinum program, which at year-end 2019 was at 3,700 agents. However, Platinum will continue to be exclusive going forward. It is available only to those agents that have strong potential to write preferred bundles and strong commitment to writing those bundles with Progressive. Scarcity will remain part of the Platinum appeal. We currently have less than 10% of Progressive agents that have Platinum. We're going to manage that number carefully as we go forward, with continued emphasis on profitable growth as key to an appointment.
We also see Platinum working as it drives consideration of the home and auto bundle and of our preferred auto more generally. I'm going to reprise a graph that we shared during prior investor relations calls, updated here through 2019. This graph compares two cohorts of agents. One group, and both of these groups of agents had Progressive Home back in 2015. One group was selected for Platinum and the other remained in the more general Progressive Home program. Of course, if we go back to the start of 2015, the more productive agents were ultimately invited to join Platinum. There is that initial delta that we see. However, after the launch of Platinum, that gap starts to widen. The momentum was initially slow as we integrated the sales teams and we looked for Platinum program infrastructure to get built out.
What we're now seeing is that multiplier effect of various initiatives that are starting to come together. Maintaining that original cohort of agents so that we can actually compare the same store sales, what we see with the solid orange line for Platinum versus the solid blue line for those non-Platinum property agents is that Platinum agents are now four and a half times more likely to bundle their home and auto. If I add preferred auto growth, these are the dashed lines that are now on the graph, you'll notice that both sets of agents have increased their preferred auto volume over time. However, once again, those Platinum agents are again producing about four times more volume per agency. The gap between these Platinum and non-Platinum lines is really that Platinum difference.
What we see when we're looking back from 2019 is that this lift has held over time. The Platinum Program itself is really about a comprehensive approach. I'll speak a little bit more about partnering with agents, both in terms of enhanced compensation, as well as making it easy for them to write bundles and to leverage a strong brand. I'll hand off to Dave, who will share how we are positioning property for long-term success with a stable, competitive offering. All of this comes together and builds our agents' confidence in our ability to serve their customers. On the compensation front, we recognize that we would have to think differently and increase compensation in order to offer a compelling alternative to preferred carriers in this space. We laid out what we call Paths to Partnership.
It allows us to compensate agents for their increased commitment while maintaining our overall low cost position. As an example of how this works, I'd like to share the story of G&G Independent Insurance. This is an agency with a smart leadership team that I had the opportunity to get to know last year. G&G joined Progressive in 2014. This was the same year that this agency opened their office in Fayetteville, Arkansas. In the beginning, they wrote mostly non-standard auto with Progressive. When we rolled out our national commission schedule as part of Paths to Partnership, they started off in the growth tier, a tier that we provide agents with an opportunity that are getting started to really start growing their books with Progressive. However, G&G's overall agency strategy was more bundle-focused, and they had a lot of preferred customers.
They were able to earn higher commissions as they wrote more preferred auto with us. Our Progressive sales representative, Ben Burleson, developed a strong relationship with the team at G&G. Ben recognized that this agency had a larger bundle opportunity and was a good fit for the Platinum Program. They moved to Platinum by May of 2018, making them one of the first 12 Platinum agents that we had in Arkansas. With this new level of partnership, G&G was at the investment tier, earning higher personal auto commissions, especially for that more preferred auto, unlocking the opportunity to earn some of our highest commissions available on the bundles they write, and they also gained access to annual policies and other marketing benefits.
Because they had moved on to the Platinum path, they were also eligible to earn a performance bonus that rewards profitable growth in bundles as well as on their entire personal lines book. G&G quickly moved up a level with Platinum, earning higher renewal commission on their bundles and a cash award. By 2019, G&G had Progressive as their number 2 carrier within their offices, gaining share quickly. The agency qualified for that Platinum Blue level, which earned it yet another cash award for reaching that goal in less than three years. In this May, we will celebrate G&G's Platinum Blue level achievement with a VIP trip, and they now have their sights set on becoming one of our Platinum 25 agents.
G&G moved from being a largely non-standard book with Progressive to becoming a Platinum Blue level partner with a preferred book, a bundled business, and written premium that has grown tenfold. What this story highlights is an outstanding agency, but the themes here play out again and again across our agents. It really underscores the power of providing different levels of partnership with compensation that reflects their commitment, as well as transparent and clear goals. These programs that we have in place now allow us to maintain our cost advantage as a broad distribution carrier, while still using targeted compensation to ensure that we encourage and that we recognize increased consideration with partner agents. Another part of the Platinum promise is ease of use, which is really part of our core value proposition for agents.
We have carried that over to the preferred bundled space through our Portfolio quoting platform. Portfolio allows agents to select multiple products to quote, whether simultaneously or to add a product during the interview. There is pre-fill available for customer, vehicle, and property information across the products that are being quoted, and the Portfolio summary page provides agents and their customers with an overview of their premium, their bundled savings, and applied discounts. Agents can add or remove a product with a click, really speeding up that sales process. Portfolio was also a win for us internally, as it was an early example of strong teamwork and deepening integration across our auto and property businesses. We launched Portfolio in September of 2018, and at the end of last year, we were live in 27 states, representing 73% of quote volume, as well as almost 120,000 users.
We are seeing the expected lift in our auto conversion. The lift in our property conversion actually is surpassing expectations, which seems to be a result of both better representation of discounts as well as easier access to the property products through this new interface. What I personally find most encouraging is that the upgrades to the underlying architecture of the platform really allowed us to better test and learn our way to continuous improvements as we respond to our agents' evolving needs and their feedback. Being a brand with a national presence that customers want or ask for is cited in our internal research as a top reason that agents look to Progressive to help grow their business. We want to ensure that agents can leverage that brand for both their home and their bundle offerings.
We know the power of the Progressive brand extends to homeowners. Research confirms readiness to consider purchase of a Progressive homeowners product, especially amongst those younger homeowners. We find that featuring home as a message performs well. Our media spend has shifted to a greater emphasis on home as part of the overall message mix. That, in turn, drives increased consumer awareness of the Progressive Home product, which makes the brand an even more powerful tool for our agents. We are leveraging that Progressive brand in our Platinum agent marketing collateral, making it easier for them to showcase both breadth of product with our co-branded materials across print, digital, as well as mass media. When we look at the overall program, Platinum is emerging as a driver for future commitment.
In a blind survey that was administered by a third party, agents indicated that Progressive is one of the companies where they place their best customers 72% of the time. When we cut that data to focus in on agents that have access to our home product, that response jumped by 10 points. Now, we have room to improve. We are still coming up short against two of the largest long-standing preferred carriers in the market. When we look at this from another angle and we ask agents where they are planning to increase business, Progressive outperforms even when compared to those long-standing preferred carriers. Once again, there is a stronger result with agents that have our home product, coming in at almost 80% intent to increase placement.
We feel good about these early years out, but we continue to learn and to look for opportunities to improve. Listening to our agents to understand their expectations and their customers' needs is critical. We do so through our agency council and in ongoing conversations with partner agents across the country. Guidance from our agents has also informed the deepening conversations across the home and the auto product teams as we evolve the bundle offering and our property product. I'll hand off to Dave to share more on that front.
Thank you, Heather. Good morning. I'll begin with a very brief overview of our financial results and then dive into some of the details. Our growth is meeting our expectations. You see, last year, our direct written premium and property grew to just over $2 billion. We're leveraging not only the Platinum Agency program that Heather just described, but Progressive's brand and marketing strength in the direct channel. The combined ratio improved by about four points last year, but is still not meeting our goals. We'll talk in a lot more detail about that. If I start with growth, I'm showing here growth in our new business sales over the last three years. The blue bar at the bottom is sales from local independent agents.
All of the growth that you see there is coming from bundles, from auto and home bundles that we're writing largely through the Platinum program. The orange bar is our direct-to-consumer business. That's pretty new to ARX, so we're seeing very high percentage growth in the direct channel. We've almost completed the conversion from the ASI brand to Progressive Home. So those of you who insure your home with Progressive, on your most recent dec page, you saw the Progressive Home logo. We've made investments in the quoting channels, so the Portfolio program for agents that Heather described is now out in 29 states.
On the direct side, we're part of the HomeQuote Explorer quoting platform, we've actually built the capability to go all the way from quote to buy the policy without having to talk to an agent in 14 states as of the end of the year. We'll continue to roll out both of those platforms in 2020. The light blue bar at the top, mostly in 2018, was a fairly large book roll. One of Progressive's partners in the Progressive Advantage Agency decided to opt out of that program. We worked with them to roll that book over to Progressive Home. When you combine all of those sources of growth, we had 61% growth in 2018.
Looking at 2019, without that book roll in the denominator, it looks like our new sales are pretty flat, we continue to have strong underlying growth in both agency and the direct-to-consumer channel. As we grow outside of ASI's original states, the mix of our business is shifting. ASI was founded as a Florida property insurer, expanded fairly quickly into the other Gulf states. If you go back to 2006 here, Florida was about 80% of ASI's business, Texas and Louisiana made up the rest. The light blue part of this graph is all other states. The expansion into the rest of the country has come to the point where Florida now only accounts for about 20% of our exposures, and those expansion states now are more than half.
That expansion also means that we are now exposed in a bigger way to new perils, especially wind and hail. If you look at the components of our loss ratio over the last few years, the orange part of these bars is wind and hail, the gray part at the top is hurricane losses, the blue part at the bottom is all other perils that the property policy covers. You see that that all other perils section has been stable and predictable. We've been very close to our pricing expectations on the all other perils section. Wind and hail, though, has gone from 10%-15% of premium to more than 30% last year. Really getting our hands around what our expectations should be for wind and hail costs and pricing appropriately is an important part of meeting our profit goals.
I'll talk in more detail later about our reinsurance program, but I just thought I'd highlight here how that results in differences between our direct and net results. The lower line here shows the direct combined ratio for ASI, now Progressive Home, over the years. The blue line at the top is the net result. In the early years, ASI was ceding a lot of premium to our reinsurance partners. We didn't have any hurricanes making landfall in those years, so we weren't ceding any losses. Our net combined ratio was higher than the direct. In the last few years, the reinsurance program has been working as intended. In years where we've had some big hurricane losses with Hurricane Irma and Hurricane Michael, the reinsurance has kept the combined ratio near 100.
It's worth noting that even in these recent years where we haven't been satisfied with our results, we've been consistent with the industry. The chart here, the blue line shows our direct loss ratio each year. The gray line shows the industry, but with the industry premium, state premium weighted based on the ASI and Progressive Home mix. In the early years there, you see with mostly Florida premium and no hurricanes, very low loss ratios. As we've expanded into other states, we should expect to see the loss ratio be higher than a no storm year in Florida, but would still been at or a little bit below the industry result. Let me talk for one minute about what we did last year to improve profitability, and then we'll transition to our plans for this year and beyond.
On the map here, the blue states, I've called the hail states. The chart on the right shows our rate increases last year. In those hail states, we took rates up almost 9% compared to just 4% in the rest of the country. We also implemented some coverage changes. The states that have the stars on them, we did two things. We started to require new customers to buy higher wind and hail deductibles. The issue there is that when fairly small hail falls on a newer, well-maintained roof, it really shouldn't do any damage. That doesn't prevent roofers from aggressively marketing to our customers, suggesting that the roofer could help them get a free roof from their insurance company.
It's clear that if a customer's roof is damaged, we want them to report that claim, and we want to pay it as quickly as possible so we can help the customer get the repair made. We have found that in cases where the customer's deductible is a little higher, they're less likely to submit a claim when there's no visible damage in response to the marketing pitch from the roofer. The other change we've made is that for roofs where the shingles are nearing the end of their useful life, we're requiring actual cash value coverage for the roof. That same fairly small hail, if it hits a shingle that's been out in the sun for 15 or 20 years, is much more likely to crack that shingle, and then that needs to be repaired or replaced.
It just doesn't make sense for us to offer full replacement cost coverage for a maintenance item on something that needs to be replaced soon. As the roof reaches sort of near the end of its life, we're asking people to have actual cash value coverage. We see evidence that both of those changes will bring the loss ratio down and mitigate the need for further rate increases. Despite those actions, in the blue states last year, we ran at 115 combined ratio, compared to an 89.5 combined ratio in the rest of the country. I think a natural question would be, well, how can you feel confident that you have addressed this problem and you can make money consistently throughout the country? Let me talk a little bit about the tools that we use to price for these perils.
Weather is very volatile, we can't use last year's experience as a good predictor for next year's claims. It wouldn't make any sense to say, "Hey, we haven't had a hurricane for three years, so we don't have to price anything for hurricane." We have pretty good models to help us understand what our likely hurricane losses will be. Until recently, we only had two models available to predict severe convective storms. Those are the big thunderstorms that cause wind and hail. I'm showing here the modeled prediction for what our annual average loss should be from wind and hail from those models over a five-year period. Here's what our actual losses looked like. The models were just not doing a good job of helping us predict what we should expect in terms of claims from wind and hail.
Fortunately, there's a new model that has become available in the last couple of years, and as we back-test that against our book of business, we feel much more comfortable that that model is going to do a good job of helping us to predict what our future loss costs will be. Model C here in the orange shows the model's prediction of what our losses would have been compared to that light blue bar, which is the actual that we paid. We're now using that new model in our pricing decisions for wind and hail. We also want to continue to grow this business while we work to improve the profitability. We are in the process of rolling out what we're calling the 4.0 version of our property product.
As we've grown outside of the Gulf states, we're collecting a lot more data in the rest of the country, we've been able to collaborate with Progressive's auto product R&D team in Cleveland to use their most sophisticated tools to help us get the price segmentation where it needs to be. Heather mentioned the feedback we've gotten from our Platinum agents. That's been very helpful as well. We've been able to expand eligibility. In some cases, our underwriting appetite was much more restrictive than their other preferred markets. We've removed some exclusions from the contract that were unusual in the market. We've broadened coverage through new endorsements.
This gives me the excuse to put the picture of the German Shepherd puppy here because we used to have a fairly long list of dog breeds that were ineligible, and as we studied further, we were able to pare that back, this little guy would now be eligible for insurance with Progressive Home. Let me talk for a minute about our reinsurance program. We maintain a very conservative reinsurance program and intend to continue that. On the catastrophe side, it's designed to provide coverage for three major hurricanes in a single year. On the far left, you see the coverage available for a first event in Florida. The way the program's structured today, we retain the first $60 million of losses from a single event. The gray part in the middle there is the Florida Hurricane Catastrophe Fund.
That's a state reinsurance fund that we're required to participate in and do. The blue section is voluntary market reinsurance that we buy. Finally, at the top, we have a catastrophe bond, an insurance-linked security. Altogether, we have coverage for an event of almost $1.8 billion in losses. Most of the blue section is reinstated automatically. If there's a big claim, we get a reinstatement. The middle bar there shows what our coverage would look like after a billion-dollar event. We would still have about $1.34 billion in coverage. We would even have, if there's a $700 million event as a second event, we would have remaining, $640 million in coverage for a third event.
Just to put that in context, Hurricane Irma is the most expensive storm we've ever encountered, and our estimate of ultimate losses from Irma is less than $400 million. We feel good about the structure of the catastrophe program. As the business continues to grow, we expect to increase our retention at the bottom slowly, and we expect to buy even more limit at the top of the program. We also have an aggregate reinsurance program. That's to cover the volatility in wind and hail that I described earlier. No individual hailstorm is likely to reach that $60 million retention. If we have a year with a lot of hailstorms, that could put pressure on our loss ratio. In previous years, we had a program that was based on a loss ratio attachment.
If our loss ratio reached the attachment point, we had a recovery from the reinsurance. We switched that this year to an aggregate catastrophe excess of loss program. In this year's program, we retain the first $375 million of catastrophe losses, and then we have coverage for up to $200 million above that. If we have a really bad wind and hail year with lots of activity, that aggregate program would kick in, and we'd have a recovery. It's important to note that change may result in a change to the monthly volatility that you see in our property results. What we're showing here, the blue line shows the property combined ratio that Progressive reported each month in 2019. The orange line shows what we would have reported if we didn't have that loss ratio-based aggregate reinsurance.
You see in January, there was a very light weather month. The combined ratio was below 80, there's no need for aggregate recovery. In February, it was a busier weather month. The combined ratio approached 120, we hadn't yet hit that year-to-date attachment point for the reinsurance. That changed in March. March was again a busy weather month, now we had a recovery on the reinsurance, our reported combined ratio in March was about 100. Each month through the rest of the year, the recovery would go up or down based on the weather. You see in April, relatively quiet. We actually reversed some of that recovery. In May and June, busier months. Again, the recovery we had on the reinsurance increased, our reported combined ratio was pretty stable.
With this year's program, we won't have a recovery on that aggregate unless we get to $375 million in total catastrophe losses. In the early months of the year, if we have a month where there's a lot of weather, we'll report a higher combined ratio, we expect the reinsurance to be available if needed if the full year results are difficult. Let me end with just a very brief description of our priorities for the property business for the year. Job number one is improving profitability. We've talked a lot about that. The second is a focus on our people and culture. Because of the fast growth in this business, over the last three years, we've gone from less than 600 people working in the property business to more than 1,200 people.
It's really important that we spend a lot of effort on coaching and career development so that those people are effective in their jobs, enjoy their jobs, and want to stay with Progressive for a long time. We have opportunities to improve processes as we grow to become more efficient and reduce cost per policy. I don't know if you noticed, on the first page, our expense ratio was down by more than two points in 2019, we see some further opportunity for efficiency gains. We want to continue to make it easy for agents and customers to quote and buy our policies, the investments in Portfolio quoting and HomeQuote Explorer buy are key there. Finally, we want to work to improve the customer experience, the focus here is on those bundled customers.
There are instances today where, for example, the billing experience is different on your auto policy compared to your home policy. We want to align those experiences so that the bundled customer has the same experience across all of our products, we think that'll result in even better customer retention, we'll keep those bundled customers for a long time. With that, we'll pause just briefly, we'll give Tricia and John an opportunity to come up for questions. Thank you.
Thanks, Dave. If you are a webcast participant, you may ask a question by pressing Ask a Question in the upper right corner of your browser window. If you are a telephone participant, you may press star one on your telephone. You may press pound to withdraw your question. After pressing star one, please be advised that you'll be momentarily blocked from hearing the live call. Additionally, there will be a brief lag between the telephone audio and the webcast audio. If you're participating through both mediums, please mute the webcast audio while asking your question. To allow us to respond to as many questions as possible, you'll be limited to one question and one follow-up question per request. To the extent you have additional questions, you will need to place your name back in the queue by selecting star one on your telephone.
Before I kick it over to Jason to take our first question from the conference call line, there's been a lot of discussion in the property casualty industry about the effects of this concept of social inflation. Tricia and John Sauerland, I thought you might want to start off by talking about how that concept of social inflation can affect particularly bodily injury severity and PIP trends.
Okay, great. I'll take a stab at that on both the personal auto side and the commercial line side. John Sauerland, if you can weigh in on anything I forgot or anything that's important to note. Why don't I have Gary Traicoff, our Chief Actuary, come up and talk about the reserve part of it, because I think that's a really important part. First and foremost, in the personal auto side, frequency is down about 3%. If you compare it to the last data point we have for the competition, it's down lower because it's flattening out. It's about zero for the competition in quarter three. That is the 12th consecutive quarter that our frequency has been down. Again, we talk about trying to attribute certain things to frequency. It's really difficult.
We would say that our mix shift to more preferred customers seems to be a part of it, but again, it's really hard to attribute any one particular thing. We'll watch that very closely. On the severity side, we're about a point different from the industry at this juncture. The typical reasons in collision and property damage component parts, actually labor rates have been increasing and we're having more frequent total losses. Those are continuing the trends we've talked about for several quarters. On the BI severity part, which we take more seriously because they have slightly longer tails. I talk a lot internally about injuries are not like fine wine. They do not get better with age. We really try to make sure we have the right file at the right rep at the right time. Here's what I would say at this point.
We see the BI trends flattening out somewhat at this point in time. This is changing, and if you're talking to some of our competition, I think we're all watching this very closely. It is flattening out, but our BI trends are up, and if you compare our incurred trends to the paid of the industry, we're about a point and a half difference, which we would assess to actuarial increases, so reserve increases. We're seeing the same aggressiveness when you think about the social inflation buzzword that Julia talked about. We're seeing aggressive attorneys. We're seeing aggressive attorneys early on in the files, and so that's really an important piece. We call it day zero before or right after we get the claim reported.
We see attorney rep rate up about 2% year-over-year on the personal auto side, and we're looking at different cohorts of coverage limits, and we're seeing about mid-single digits in every cohort, and they're slightly different depending on the-- Think of 5,100, 2550, et cetera. This is an anecdotal piece, but some of our CRM reps have said that they are getting calls from someone other than the named insured to assess what the limits are. I remember being in a claims branch for 15 years and getting those calls. Oftentimes they were from plaintiff attorneys. This is anecdotal, but am I more anxious to take a file that's a 5,100 than a 2550 file? Those are just some things that are anecdotal.
A new piece of information that we have from a vendor that we work with shows that attorney media spend is up 10% when you compare the fourth quarter of 2019 to the fourth quarter of 2018. I could have anecdotally told you that from my travels, and you turn on the TV in any state you go to, any city you go to, look at billboards. We think that that's getting more aggressive. We have to watch that really carefully. I assume that many of you will ask because I had referenced a handful of states a couple of quarters ago that we're watching closely. There's about five states that we've been watching from BI severity trend. Four out of five of those have shown a decline in severity in the fourth quarter compared to full year of 2019.
Three out of five of those states are still higher than countrywide. We're watching it closely. Right now it appears to be flattening, but again, that can change at any given time. We are very surgical in pricing each line coverage, each state, channel, product, all of those things, so we'll keep on top of that. Clearly, our margins, we believe, are really strong, and we believe we're conservative from the reserving side. On the commercial side, we are down 4% from a frequency perspective, up about 19% from a severity perspective, and about a couple of those points are reserve increases, actuarial strengthening. I would step back and say one thing. Every BMT in our commercial lines organization was at or below our target. Very successful 2019. Into January, we were sub 90 combined ratio.
We are very strong on the commercial side. We also are no stranger to being able to react to trends. If you recall, in 2016, we bumped up against our 96, and we knew part of that was commercial. We immediately increased rates, knowing it takes a little bit longer because the majority of those policies are annual. Still continued to take rate in 2017 and 2018. 2019, we saw the competition was also taking rate during those time frames, but in the first half, that started to diminish a little bit. It's picked up at the latter half of 2019. I would say for Progressive, we took less than a point, a 1.5% rate increase in 2019. We will likely be a little bit more aggressive in 2020 as we see specific states and loss trends increase.
In terms of the 17 or 19% BI trend, part of that too, we are having a mix shift to our for-hire transportation, which are higher severity. That's one other piece. When I talk to John Barbagallo, he sees three states very specifically where we see loss trends accelerating. We either already have filed or in place rate increases. In addition to that, we have three states, those same three states, we have multiple variations of underwriting restrictions. As an example, we have a 9.6% rate increase going in May in California. We'll watch those closely. Again, when we think of growth and profitability, if you have to make a choice, which we never want to, it's always going to be profit. One, I feel really great about our solid results in commercial, especially compared to the industry.
We will get on anything we don't see quickly and make sure that we meet our profit targets. Do you want to add anything?
If we're going to ask Gary to come and talk about reserves, I'll fill in while he makes his way with two thoughts. One, it's important to take a little step back when you're thinking about trends, frequency, and severity, and look at the longer-term trends. We think in aggregate, our severity trends are normally very consistent with the industry, and frequency trends, we actually have been enjoying bigger drops in frequency more recently over the past three years, actually, than the competition. Tricia mentioned some of that is due to writing more preferred mix of business. I would offer we also believe that it is due to more robust underwriting we've put in our upfront process in binding new business. We think that's had a lot of great outcomes in terms of avoiding risks whose intent is not to insure but to defraud.
Longer-term severity, sort of where the industry is, frequency better than the industry is. I would also offer, where Tricia was mentioning on the commercial side, matching price to risk, we are also very agile in the personal lines side, and we'll certainly continue to make sure we are matching prices that we perceive should rise, at least on the liability side, due to the trends with risk as fast as we can.
Gary, what are your thoughts?
Great. Well, this is Gary Trajkov, Chief Actuary. Hello, everybody. Tricia and John gave a great overview and description. With respect to reserves, as you know, we developed unfavorably last year about $232 million, which was 0.6 of a point on the combined ratio. That development was primarily related to the increasing injury severity trends that we were seeing and led to unfavorable case development. We recognized that early in 2019, and over the course of the last three quarters, we increased reserves from actuarial changes, roughly about $60 million. We ended up going up about $186 million during 2019. In addition to that, our claims adjusters continued to strengthen reserves through natural movements as well.
When we look at it overall with the changes that we took, we now on an accident year basis are up about 12% for commercial auto year-over-year and 6% for personal auto. That would be loss and LAE for liability, which you probably noted in the annual report. Of course, LAE is a little bit flatter, so on an indemnity side, we're a little bit north of that. When we think about how development is occurring recently, last year over the last two quarters, we did develop slightly unfavorably about $20 million. Of that $230 million that we saw come through, 210 of it was in the first half of the year and a much smaller amount during the second half of the year. In January, we ended up developing unfavorably about $78 million, which was a little over two points on the combined ratio.
In January, the development was really related to some other areas. We primarily developed unfavorably due to December claims that were reported in January. When we look at our injury case reserve development between personal and commercial auto combined, we actually developed slightly favorably in January. That's just one month, so it's hard to say how the future will go, and I definitely can't predict how we see development and for the year or the changes that we take during the year. As you know, as the year plays out, primarily the development we see on the injury case reserves are a main driver of what we end up seeing.
Great. Thanks, Gary.
Okay. Thank you.
Great. Jason, now can you please take the first question from the conference call line?
Certainly. Your first question comes from the line of Mike Zaremski from Credit Suisse. Your line is open.
Hey, thank you for all the details. My first question is on any potential impact from the current situation with the coronavirus. The New York Times has come out and said that they're seeing just recently ad spend fall fairly materially across the board by as much as 25%. I'm curious if you think Progressive should in the near term or is this part of that? Also, are you seeing any impact maybe from your telematics drivers on lower frequencies if people are maybe working from home?
Thanks.
Mike, that's a great question. I'll start with the ad spend. Right now we're going to continue to spend. This is a prime time of the year when people are buying insurance. We're getting into that season, we'll continue to spend. That we have some flexibility in. Again, whether you drive a little bit or a lot, you still are required to have auto insurance. Our intentions will be to spend as long as we feel sufficient. Again, we'll have to be nimble because all of this, as you know, is ever-changing. The great question on the UBI. With the recent deaths in Washington, we asked the UBI team just to take a look at UBI vehicle miles driven or traveled by week in January and February this year compared to the prior two years. We are not quite seeing a difference.
Again, that's very little data, that tells us we haven't seen it yet. Again, now that we'll look at it weekly, we can start to see that. We'll look at it across the country where we can. We'll be able to understand pretty quickly. If you go back to something like the financial crisis, I was running claims at the time, we saw frequency drop really quickly, we'll have some good insight. We get our frequency data on a daily basis, we'll understand very quickly where we're at. From a vendor perspective, we always think of the concerns around auto parts that are possibly made in China. We had our property process team talk to all of our OE vendors, the percentage of OE that we use on our vehicles, the percentage they get from China, et cetera.
For the most part, with the exception of one OE, we feel like there's low risk at this time, and even with that partner, they have an inventory. Again, it's always those first and second order effects. It could be that more cars are totaled because you can't get parts, and then there's used car parts. We're going to keep watching that. From an internal perspective, we already have over 25% of our people working from home. We have had many team meetings. We're having a tabletop pandemic exercise tomorrow, I believe. Our chief HR leader, Lori Niederst, had a meeting yesterday with our chief medical officer talking about the same things that most companies are talking about in terms of non-essential travel and what to do if you're coming from a country that's been affected. Right now we aren't seeing any effect.
Again, this is such a moving target that we have a lot of data points that we're going to be looking at literally on a daily basis to understand how it will affect possibly our frequency.
Okay, great. Lastly, just to follow up to the actuarial comments at the end of the prepared remarks. I believe you said that January's reserve development was worse than expected, was fairly material. Are you saying that that was mostly due to December claims that maybe people just didn't make due to the holidays? You said in January, ex that was actually favorable. This implies that just last year's loss ratio was worse and that on a forward-looking basis, things looked a little bit better.
Yeah. I'll have Gary talk about that. Yeah, partly it was the December losses late reported.
Sure. Great question. In January, we were about of the $78 million, pretty much all of that really related to December claims that were reported in January. We look at it throughout the year, right? When we look at January claims that come in in February or February claims that come in in March, et cetera, some months were high, some months were low. It's really noticeable in the first month because it's prior year coming in. If we excluded that, those late reports that came through, our development pretty much was right at zero. In addition to that, if we looked at just injury case reserves, which was the primary driver of the unfavorable development last year, we actually were slightly, very close, but slightly favorable in January.
Those claims, as they paid out, came in a little bit below the initial reserves we had set. Again, not an indicator that that's how the year ends up, but that's what we had in January.
Gary, over the years you've shown us that that one data point in January, how it evolves is very different in every year. You showed us a comparison of three or four years. What I would say, Mike, is one data point, strengthen it. We have oftentimes the December late reports, I would say we are all over this and feel good about where we're at. Of course, we'll react quickly should we need to strengthen.
I could put it in perhaps simpler terms in combined ratio points. Simply because we're turning the page, what Gary was saying, we're going to see some losses we categorize as prior year every January. We had 2.2 points of prior year losses in January this year. Last year, we started out the year with 4.8 points. That was a lot higher than we've normally seen. As Tricia was mentioning, generally speaking, in January you're going to see some prior year development and the 2.2 doesn't concern us at all.
Thank you.
Okay, great.
See you again maybe possibly.
Jason, we'll take the next caller from the conference call line, please.
Certainly. Your next question comes from the line of Elyse Greenspan from Wells Fargo. Your line is open.
Hi. Yes, I was hoping, Tricia, you could provide a little bit more info on your outlook on the personal auto rating environment. It sounds like from some of your introductory comments that you continue probably to expect on pretty modest rate movements, I guess, throughout the majority of your book for 2020. Has anything changed or maybe in some specific states where you're taking a little bit more rate?
Great question, Elyse. We'll continue to look at that and as it evolves, it's really hard to have that crystal ball. Last year, and most of the industry, most, some took some overarching rate decreases. People were taking a little bit of decreases. People want to start to grow a little bit. We're seeing less of that. The industry is getting closer to 1% at this juncture. We are very surgical in each state, like we said, each channel, each product, and we'll react to that as necessary. We feel really good where we're at in terms of our profit margin. I talk to Pat Callahan, our personal lines leader, all the time on specific states and what we need to do to strengthen it. Again, we don't want to get behind.
It's really important for us to have stable rates for our consumers, we're going to take that 1% or 2% to make sure we reach our target margins. We feel really good, specifically on the direct side of the new business coming in. We have new business targets as well, we feel really good about it at this juncture. Again, I feel like we're really nimble when we need to be, should we need rate, but we feel good at this point, again, point here, point there, depending on what we're seeing in specific states.
Okay, my second question, could you just provide an update on the small commercial side of things? You guys were rolling out some products in one state, the expectation was to maybe expand into more states. Can you just provide an update on where those initiatives stand today, how you're thinking about additional steps and rollouts throughout the rest of 2020?
Absolutely. Mid-2019, we rolled out Ohio, literally small, five agents trying to figure out was the product ease of use, et cetera. Got the thumbs up, got some feedback on pricing. Rolled out to Ohio and three more states in 2019. Since then, we've rolled out two additional states in January, two more in February. We expect to roll out two more in March. Total for 2020 should be in 15 states. We've got 2,000 agents selling that small business, we are really excited about the momentum. This was something, as we thought about the three horizons, you think about investing before you need to make sure you have that enduring business. We started thinking about this a couple of years ago, we were really excited, actually, about small business, both on the agency side and through our Business Quote Explorer.
We'll have the Progressive product hopefully on the Business Quote Explorer sometime in 2020. We have many different unaffiliated carriers, partners that we work with, we continue to be able to give the small business owners what they need. I would say the one word about small business would be momentum. I feel really good about where we are. Actually really good about where we are with a lot of the topics that John Barbagallo and Karen Bailo went over a couple of quarters ago. We rolled out our Small Fleet program to 49 states, the conversion has increased fourfold. Obviously, our relationship with both Uber and Lyft and the TNC has increased, we're excited about that. Just across the board, I feel great. Our Smart Haul program is showing great conversion, great take rate, I should say.
That's our UBI in commercial, in fact, on the agency side, where the customer is eligible, the take rate's 25%. I would say commercial's firing on all cylinders, small business, and everywhere. Do you agree?
Absolutely, and I share that excitement. Just for clarification for all viewers, when we're talking about the rollout here, we're talking about business owners' policies and general liability. As Tricia mentioned, we got into four states in 2019. We've elevated two year to date, and we expect to actually add about 15 states for this year. Ending the year maybe around a little over 20 states. Again, this is intended to vastly broaden our addressable market for commercial lines. We've been number one in commercial auto for a number of years now, and this opens up a marketplace that is probably two, perhaps even three times the size of commercial auto for us. Very excited about that growth, as well as the plethora of other great things we have going on in commercial lines.
When you think of that bundle customer with BOP and GL, and then you think across our channels as well, there are many small business owners that actually also have our auto and home. As we think about that, we really think about the household economics going forward, as that's really what's exciting as well. Thanks, Elyse.
Your next question comes from the line of Michael Phillips from Morgan Stanley. Your line is open.
Thank you. Good morning. I guess a bit of a continuation from that last question in a different angle. A large part of your earlier comments on the slide presentation this morning was on the agency channel, and that's where the Robinsons live. With the focus there because of that, can you talk about any maybe incremental help that that does, more focus on the agency channel that helps you with your commercial lines offerings?
Many commercial, whether it's small business or commercial auto, actually go through the agency channel. It's a little bit more of a complicated product. That's actually a much higher percentage than would go on the direct side, although we believe at some point we want broad coverage, or actually now we want broad coverage for everything. I believe, as I talk to agents, there's some agents that are only personal lines, some are more commercial. There's many, especially large agencies we work with, that are both. For them to have access to all the products they need for that customer, whether they have a small business and their auto and home, is really a great umbrella for all of them to serve their customers, and that's what they want to be able to do.
I think it's really important in the agency channel because it is still a little bit more complicated. If you think of a person who's opening their first business, they want to make sure they're protected. They want to be educated. That is nicely done through the agents. We're very bullish on that as well.
Okay, great. Thanks. I guess, back to frequency on the personal auto side, how does that vary by, I guess, age of car and model year? Maybe the reason to think about that is there a continuation of continued frequency to the extent that it's maybe more of the recent car years versus the prior ones?
Well, we look at it mostly based on customer. From a preferred to a non-standard, who are more likely to have accidents, or we look at it in terms of the demographics of, are you a mature driver or are you just starting to drive? That's how we look at frequency, rather than types of car. What I said is it's really hard to attribute very specifically to frequency, but we do believe a piece of it is more of our preferred customer who likely have less accidents.
Yeah. As Tricia was saying, diagnosing exactly where the frequency is driven, meaning by the driver of the vehicle, the environment, all that, it is very difficult. That said, if you're focusing on model year vehicles, certainly newer model years are driven more miles than older vehicles. We've been growing a lot, and we've actually been increasing our share of those newer models as we write more and more preferred business. The fact that our frequency is down in the same time period of the trend that I just described there makes us pretty confident that we're writing the right preferred business.
Your next question comes from the line of Yaron Kinar from Goldman Sachs. Your line is open.
Good morning, everybody. My first question goes to the partnership with the ride sharing companies. Is it fair to think of the incentives as not necessarily fully aligned, namely the ride sharing companies I would think are very focused on growth, would probably be interested in settling claims as quickly as possible, maybe not necessarily pushing back as much, when you guys may think it is necessary. If that line of thinking is correct, I guess, how do you manage that risk or that misalignment of incentives?
Actually that hasn't been an issue. We fully handle the claims. They're completely done in-house. The great part about Progressive that I've always felt, especially having my upbringing in claims, is we really have never even differentiated between an insured and claimant. Every consumer is a possible customer, et cetera, and so we settle fair and accurately. We don't get pushback from them. I haven't heard anything about that. What they want is somebody out there getting their driver's car back on the road so they can make a living, and if there's injuries, making sure we're fair and settle those. We haven't had that issue. I think they look for partners that have a world-renowned claims organization like we do.
We have feet in the street because we have local presence. It's really worked for both Uber and Lyft, and all the feedback has been that we do a really great job in that. That's how I see it in terms of, they want to have the claims handled by somebody who has a history of doing the right thing from an indemnity perspective and that are also cost conscious from an LAE perspective.
From a financial perspective, I'll point out that in both of our ride sharing relationships, there's a quota share agreement. In both of those cases, those companies have captive reinsurers that are part of their organization, and we are ceding premiums losses. They are sharing in the financial results that we're experiencing with all their drivers.
Okay. My second question just goes to bodily injury severity in broader terms. Can you maybe talk about what accident years you saw the increase in bodily severity coming from, both in personal lines and in commercial?
You probably have to help me on this one. I would say if I had to guess while he's looking it up or you can, more like 2017, 2018, where we're starting to see it develop. I can't say for certain without looking that up. Again, those trends do develop a little bit more over time. I know that off the top of my head, that's what I would say.
Of the $232 million of prior year development, approximately $131 million was from 2018, $73 million from 2017, and the remainder from 2016 and prior. We detail all that in our annual report.
I was like, I know I've read that, I'm going to guess, luckily I was right. Great.
Okay. Those ratios are relatively, the distribution between those accident years is similar in commercial lines and personal lines?
I think it's safe to assume that. I actually don't have those numbers to quote for you.
Okay.
Generally speaking, older accident years have already developed previously, and by a large part they have settled. We also provide in the annual report, loss triangles where you can see where we picked, if you will, the loss reserves at the end of the respective year and how that develops over time. You can also see the % of those claims that have been paid, and obviously on physical damage claims, those get paid very quickly. On a bodily injury, you can see that development. Especially on the personal side, those bodily injury claims, while certainly take longer to settle than fixing a car, they develop fairly rapidly. Commercial lines, a little longer, but you can see all that in the annual report. It's safe to assume that prior year development is predominantly from the most recent year.
Thank you.
Your next question comes from the line of Gary Ransom from Dowling & Partners. Your line is open.
Yes, good morning. You mentioned briefly during the presentation about the direct side of the homeowners business. Can you talk a little bit more about why that's growing more rapidly? What kind of customers, whether it's bundled customers that are coming in on that side as well, or any other comments you might have on the direct growth?
We've grown, I think I've talked previously, and we're going to have one of the upcoming quarterly webcasts, sort of a spotlight on our Progressive Advantage Agency. In our Progressive Advantage Agency, we have Progressive Home, along with many other unaffiliated carriers. That way, we're able to really have broad coverage for the customers that come in. Maybe Progressive Home doesn't want that risk, but another company does. We have a very low DNQ rate in there, so we're able to bring that in-house. We've grown our Progressive Advantage Agency substantially in the last three or four years, that is one way where customers want to come in.
In addition, we have HomeQuote Explorer that we developed a couple of years ago, where you can go online, we also have Progressive Home as well as several other unaffiliated carriers, we have a buy button with that in 14 states. When you're able to go on and actually purchase, I think it's really important, we'll continue to roll out more and more states with that. I think it really is customer preference, that goes to our strategic pillar of broad coverage. If you feel comfortable, the great part about HomeQuote Explorer is that we're able to gather a lot of information from publicly available data to make the quote really easy.
Especially if it's a pretty simple basic home with things that we're able to get, they can get it done really quickly, some people don't necessarily want to go through an agent. We've got both areas growing rapidly, but it's really great in the direct side because a lot of people want to go either on the phone or online. Great.
Can I just expand on that question on the-
Sure.
Moving it into commercial as well on the commercial side, it also, you're growing more rapidly on the direct side, as it says in your K than in the agency side. Is there any characteristics of the customers there that are bringing that growth stronger?
Well, yeah, it's also on the base. BQX is fairly new as well. The customers are the similar type customers. It is more complicated. I wrote in my letter that I sat with a BQX rep, it gets really complicated when they start adding different coverages that they want. We feel like right now that we can accommodate about 70% of the small businesses, that's why we're going to continue to have more and more partners and then ultimately have the Progressive BOP GL in our BQX.
I would say BQX is less mature than HQX, HomeQuote Explorer, they're doing similar things that we did several years ago on the personal auto side, that is build an in-house agency, utilize partners so we can cover many different types of small business owners with different products that we may or may not write on our paper. I think that's the really important part. It's a similar thing. We want that bundle. Where we believe we'll have longer tenure commercial auto partners is if we have more of their commercial needs. Same thing on the personal line side.
The one thing I'd offer to add to that, Gary, when you're looking at the premium mix numbers, direct versus agency for commercial lines, you should be aware that we categorize the Rideshare Partners business as direct. Those obviously are pretty significant premium relationships, as we add those and add states, you're going to see that growth.
Yes.
Thank you for that.
All right.
Your next question comes from the line of Meyer Shields from KBW. Your line is open.
Great, thanks. I just want to start by thanking you for the enhanced capacity disclosure and maybe more importantly, for the responsiveness. Tremendously welcome.
Our pleasure.
This is a bit of a weedy question for Gary. When I look at the triangles in the 10-K, the H to H factors for auto liability agency, direct and commercial, they're all speeding up, and I was wondering what that actually reflects?
Yeah, sure. Gary, be up in a second, Meyer.
Hi, Meyer. Now when you were relating to that, are you looking at the paid or the incurred?
Both. Paid.
On the paid side. Yeah. On the paid side, there's a couple of things going on, right? There are definitely some states where we are seeing a speed up in closure rates, particularly 30 days, 60 days, 90 days coming in as well. On the incurred side, you may notice some changes. What we have seen is our adjusters, we feel they're recognizing larger claims quicker, and so they are recognizing those claims, and we're seeing the numbers come up quicker, which would mean theoretically then we would see lower development factors on the paid and incurred later in the triangle. Right? That's some of the subjectivity that's coming through now where we're seeing that speed up early. The question is, how much of that do we think will materialize later on, or it backs off as it develops to ultimates?
Okay, no, that's very helpful.
Okay.
Thanks. Second question, this is completely unrelated. Just I want to understand the thought process of raising deductibles in the health state and changing the coverage instead of pricing for the specific option that the customer would want
part of it is if you have a 20-year-old roof, you couldn't have rates enough to cover that if someone has a hailstorm, and we replace that fully. We try to create different coverages that put skin in the game. It's been really difficult with vendors out there. We'll go through the storm process, and you can see it as they develop. They'll go through, and you're knocking on the door, and you're making sure, "Don't you need a new roof? Your insurance company should pay for that." We're trying to always pay fairly, always do the right thing for our insureds, but have some skin in the game that you don't just replace your roof every single time there's a hailstorm when there isn't damage, or when there's slight damage that isn't actually changing the structure of the roof product. We'll see how it goes.
To price to a hail state would be no growth, I believe. We're trying to be creative in our product development.
Okay, great. Thank you so much.
Thanks.
Great. I'm actually going to take a question from the webcast. It's about policy life expectancy, an important topic we haven't really talked about yet, and I got a couple questions about it. Particularly in direct, what is causing the decline in policy life expectancy, both on a 12-month and three-month basis?
Well, a couple of things. We had gone over a process that we changed a while back, occasionally we have that happen. There's another one that we're doing that I don't want to talk about for competitive reasons that will actually negatively affect PLE, we think it's the right thing to do to have the right customers on the book that are actually we can make money on. Also, it's been very competitive. Rates have been really stable, there's a lot of advertising out there, it's really easy to change. There's a lot of consumers that are just price sensitive, they shop all the time. If they're going to shop, the likelihood they can find a lower rate with us or some of our competition.
I will say, of course, it's one data point, PLE has lagged, that the December development has actually increased in both the PLE in the three-month and the 12-month. Again, I don't want to say that that's the future. We look at that as possibility. We look at PLE very specifically with nature, nurture, and price. Nature is our mix of business. Obviously, we want more of the preferred business. Nurture is how can we take care of our customers. We are investing a lot in the CRM organization around sort of n equals one personalization. How can we be there for you particular, communicate with you in the way you want. Then, of course, price is the competitive landscape and the ease of going back and forth. Those three things we look at from PLE.
We continue to have a team that works on PLE. I'm the executive sponsor for that, we've changed leadership to make sure we look at all different angles. In the spirit of question everything, we look at PLE, we'll continue to look at that externally overall, I think there are some cohorts that we believe that we can increase PLE more substantially than others. An example would be there are some Sams that we call, some not all, that are just inconsistently insured, we love them. That's how Progressive was born, as long as we can make our target margin, we're great, they might go. Will we ever really make that go from X to Y? We'll always treat them nicely, occasionally they become Robinsons. They're the other cohorts that we say, "What are other things?
Why are you leaving? What are other things that you need from us on the nature, nurture or price?" That's what we're working on, internally looking at PLE from very different cohorts. PLE X Sam, PLE when you have auto, home, umbrella, all those things, we're starting to really gain some traction on how we think about household PLE.
Great. Unfortunately, we've run out of time today, I'm going to kick it back to Jason for the closing scripts. Thanks for joining us.
That concludes The Progressive Corporation's fourth quarter investor event. Information about a replay of the event will be available on the investor relations section of Progressive's website for the next year. You may now disconnect.