Good afternoon, welcome certainly to the guests from out of town. Hope you appreciate our great weather. I'm going to spend the afternoon talking about growth opportunities for Progressive and frankly, a positioning of the company. Let's make sure that we all respect this during the course of the afternoon. Before I get started, I'd like to give you at least my impression of the health of the company. A little bit of the state of the union, taking you through a few things that I think are central to Progressive. Just quick checkpoint, skills, where are we, then we'll go into the positioning. To me, central to everything we do is matching price with the risk that we take. Some of you all have seen this diagram before. Don't worry about calibration here.
Where we have need for calibration, we'll do that a little later in the presentation. This is about the macro ideas. Segmentation, so central to everything we do. The black line is central average pricing, no segmentation at all. The gold line, perfection. Our quest, always to be pushing out closer to that gold line, most importantly, to be further out than our competitors. Simple as that. We'll never get all the way to the gold line. Hate to admit that, it's true. 3 things that are really helping drive our segmentation right now in and above the things that we've done over the years. Telemetry, Snapshot if you like, being able to use that changing technology curve of automobiles and starting to have them talk to us and better understand the driving behavior, incredibly powerful. You know most about that's one of the key dimensions.
The second is our access now and wealth or becoming a wealth of data around more preferred customers. We're getting a better customer data set on preferred behaviors that really help us out in a sector that we haven't been as strong in from a segmentation point of view. Third, really just the ability to validate what was heretofore many other variables that weren't easily validated with external data sets. Those 3 effects are just things I'd highlight in terms of our continuing to move out this curve. We incorporate those features into product models. We're always doing R&D. It's like painting the Golden Gate Bridge. We're just continuously doing it, starting at one end, moving to the other. At any given time, we may have different variations out in the marketplace, always trying to get our most recent ideas out.
We don't tend to name our products perhaps the same way as some other companies, we're always doing that. Price to risk, I feel great about. For me, you can't build on anything unless you've got this right. 2nd, most of you will have seen this curve, hopefully many times. It's one of my favorite models of all times. It means nothing to a lot of people, it means a ton to us. The Y-axis, total loss cost plus LAE as a function of LAE. We have spent years and years driving down the curve to the lowest possible combination. Secondarily, changing the shape of the curve, pushing that gold curve to the blue curve. It is impossible to give you a sense of why we think we're succeeding there. I will tell you today that we have the best quality of claims.
I've said this many times, it's been true every time, the best quality of claims relative to our own measure ever recorded. While it wasn't a primary objective, we are now doing it at the lowest loss adjustment expense in the industry. A really phenomenal textbook case of moving this curve and trying to find the many thousands of things that actually make up a claims process which seems so obscure if you're not really vested in the business. Happy with that. Third, retention. One where I'm never going to be very happy because there's always opportunity. We're going to show this curve or curves like this a lot today, it's an important one. Normally, we like short tail and relatively low development, certainly in our reserving. With retention, very happy to have long tails and favorable development.
We've been moving the curve out from the gold to the blue, you'll see some of that quantification a little bit later, by fixing what I've called in the past friendly fire. Trying to find ways to eliminate things that we do that cause people to leave us and break the relationship that we started. We've done a good job on that. There will be more to do, we are a much better company today than we were even a few years ago at giving people all the reasons they need to stay, or eliminating the reasons they need to go. Today, we're going to talk a lot about changing the shape of this curve. In fact, a simple way to think about today is thicker tail. Thicker or fatter tail. More people staying with us longer in their consumer and insurance career or tenure. Costs.
We highlighted this last year. Non-loss, non-acquisition cost. Not quite so visible for you in some of the public reporting. The key point here is we have made a lot of progress on cost as well. The curves are not going to show dramatic change. Scale has helped a little bit. We control our IT cost to make sure they never go up faster than our own premium, therefore, we do get a little scale on that. We've been able to get a much greater leverage from what I'll call the digitally responsive consumer. Much more of our servicing, we'll refer to it during the course of the day as CRM, customer relationship management, or servicing customers, has been advantaged by the fact that so many consumers today are doing so much for themselves.
We're in a position to capitalize on that, to give good applications, whether it's on the internet or mobile devices, that our cost of servicing per customer is going down. Our leverage headcount to policies in force has been working quite favorably for us. That extends into real estate, it extends into the many thousands of things that go into cost management. We have a very empowered workforce that often sees their opportunity to influence gain share very much in this way. We're at a very very strong place, recognize this only gets better with long tenure. Increased leverage there could be quite significant. We would say, as best as we can do it, when we normalize for life expectancy of customer sets, we have very comparable, if not as low a cost as anyone in the industry if it's normalized.
Managed media and channel equity. I just want to make a point here. Let's take the gold curve for a second. I don't think it'll be a mystery to anyone in the room to suggest that the costs of direct acquisition, especially in a startup situation, are so astronomical relative to your own premium. You don't have the book of renewals. I conveniently cut this graph off at 1998, largely because I was responsible for it before 1998, and it was considerably higher. Yeah, it's sort of meant to be a little bit of a chuckle, but I do remember setting myself a goal to have my expense ratio lower than my loss ratio. They promoted me instead. The blue line.
The blue line's a little confusing because at the time I show on the graph here, we actually had three commissions in the marketplace, the blended value is the solid line, it won't take much imagination for you to remember when 15% or thereabouts was the clearing cost for agent commission. The point I want to make in this relatively simple diagram is the convergence of our cost of acquisition. That is hugely important to me. We've been able to get our agent cost and our direct cost to the point that there is no fundamental arbitrage between acquisition costs between channels. That is ensuring, for us, the viability of the agency channel.
There's no question, like anybody else, an agent might like more commission rather than less, here we've been able to truly understand, through our work in direct and agency, what the true market clearing cost can be for cost per sale. We've been able to achieve that in both channels. I think this speaks a lot to the viability of both channels going forward. There's no question, with greater leverage in the direct channel, with lifetime earned premium or life expectancy of customers extending, yes, there's leverage there versus the variable cost. On a simple income statement basis right now, they're very comparable costs. People and money. To me, our business is relatively simple. It is people and money. As much as we could find ways to describe our culture, I'm still left very short to describe our culture.
What I've put up here is very simply a blue line that tells you that when we interview people as we do every year, what's your likelihood of being at Progressive and enthusiasm of being at Progressive a year from now? It's in the 90th percentile range and pretty constant. Of those that are not there, it's really a neutral. Presumably, they've got plans to do something else. We have a great culture. Brown line is simply a turnover rate of non-entry-level jobs that's sort of four or five kind of range. I actually hadn't planned to say this, but I saw a study today by, I think it was MarketWatch, where that's I don't know how they did the study. Ranked us in the top 10 cultures of companies that they surveyed. In fact, I think if I remember correctly, we were ranked number one.
I don't really get into studies like that, but as I've gone further in my career, I recognize that a strong culture that allows you to implement your ideas is so critically important and yet possibly so devalued when we analyze a company. It's not good enough to know what to do. You've got to be able to get it done. I'm really proud of being part of the culture of Progressive that's got a winning attitude, a high meritocracy, a measured output, yet decency in all dealings. Only because we have these kind of attitudes can we do the kinds of things that I think we've done over the years. Our brand.
It wasn't that long ago that I would address a group like this, let's say circa 2001, 2002, 2003, I think there was a fair amount of doubt in the audience as to whether Progressive could actually add to an established skill set of segmentation, claims management, and technology. Could we really add a marketing skill that was going to be sort of comparable to industry best standards? I think it was fair to have those levels of doubt. My title here, elusive to somewhat interesting to an asset, is sort of how I feel. There was definitely a time where our brand felt elusive. We had trouble matching up what we thought was really the ethos of the company with the messages we wanted consumers to understand about us and the products we had to deliver.
We went through a very long period of sort of searching for that. For the last several years, I think we've really found a combination and now have an established brand to build on. Much of what we talk about the rest of the day wouldn't matter if we didn't have this as a foundation. Capital management. Really not much to say here other than what we say is what we do. Hopefully, we've been very clear about our layers of capital. Our three layers of capital, and specifically how we'll use the third layer if and when appropriate. The four circles there represent times where we've paid a variable dividend, which is working well for us. We're very happy with that.
We've done some share repurchases, even though that three out of the four represent a special dividend, I would tell you that a special dividend is still exactly that, special. We'll use it when it makes sense for us to use it. The big takeaway is that we've been very clear about what we say our capital management approaches will be, and hopefully you recognize that we actually do that. Just rounding out the comments that I want to make about sort of the health of the company. I want to give just a very brief tutorial on what I call lifetime earned premium. Clearly, it seems very simple, should be straightforward.
Yes, the title gives most of the information, but it does get a lot trickier as you realize that these things change over time with rate increases, with different actions to increase people's life expectancy, and just keep that in the back of your mind. For now, a simple model will work well. This notion of lifetime earned premium, one acronym we'll use because we use it enough today, LEP. I'd like to just take you through a quick example so we don't have any ambiguity. Some of you might remember, and if you don't, we're going to do a little refresher tutorial in a few minutes on some classes of customers. For right now, it doesn't really matter, but I'm going to say Sam is one of our customers who has a lifetime earned premium of $1,000.
Sam actually consumes his entire lifetime earned premium in one calendar year. At the end of that calendar year, there is no unrealized or unearned lifetime earned premium. It's fully earned. That is very much a fill the bucket every year kind of customer. Let's move to a customer we'll call Diane for now. When we take Diane in, based on our expectancy of her tenure with us, we think she'll have $2,500 of lifetime earned premium. She too happens to expend $1,000 of it in an annual year, or the first year. At the end of that first year, there's $1,500 that is still unearned. Now, if I took just a simple ratio, hang with me here a little bit, of the unearned over the incepted earned, I'd have 43. 43%.
It's a little bit of the opportunity or the recurring revenue that hasn't yet been recognized. One more example, just to round it out. Robinsons, a customer set that will stay with us even longer, and I'll give them a $10,000 lifetime earned premium. They too, for simple math, consumed $1,000 in the first calendar year, leaving $9,000 yet to be consumed. Same ratio concept of unearned over the initial incepted lifetime earned premium, 78%. All I want to do here is give you a sense of the difference, not an accounting measure, but it's a very interesting book measure of the composition of the book. I could have the same income statement with three Sams with zero carryover versus, in this case, Sam, Diane, and the Robinsons with a significant amount of carryover. That's what I sort of call, in my terms, the invisible balance sheet.
At Progressive, we clearly know that many of our competitors are positioned quite differently than us. Their history gives them an opportunity to have a great deal, a much higher ratio. We're also moving on this ratio, and a lot of what we're going to talk about today, it's important to get that notion through and think about some parts of what we'll do will materialize relatively quickly, and some parts will just be a cumulative build over a long period of time. Those are the primary points that I'd like to make. On segmentation, on claims, on cost management, on culture of the company, on capital management, I feel very good. I think I have the basis to feel very good about that. Great job to the company. Now what we'd like to do is sort of set up for the next hour.
To do that, I'm going to take you on a little bit of a history lesson of Progressive, and at least from a business perspective, give you what I think were key eras of Progressive. To do that, I'm going to use a backdrop of 1985 premiums through to last year. It'll work just fine. Clearly, there was life of Progressive long before 1985, all the way back to 1937. I didn't choose to include that because the bows got very small, and I wasn't here, which was probably the main reason. The first era of Progressive, I'm going to call it the manufacturer and wholesaler era. Again, I wouldn't worry too much about my delineation of specific years where they start and end. That's really not the point here. Within reasonable parameters, that should get us there.
Let's, in its crudest form, what was the manufacturing and wholesaling era of Progressive? Others gave us what they didn't want. In many respects, we built a skill set, most of what I've just covered. We built a skill set based on taking on what others didn't want. The defining nature of that first era was, for me, characterized by there is no bad risk, just a bad price. We didn't have Snapshot then. Our make and model segmentation in its day was just as breakthrough. We built a claims organization because we fundamentally understood we couldn't perfect a model of cost matching without controlling our claims organization ourselves. We built out a national footprint of agents.
As crude as it would seem by today's standards, we had some agent-facing technology and even some consumer-facing technology that gave us the reputation of being a technology leader in the industry. We built a product management system that used, even by today's standards, some pretty reasonable data systems and management reporting systems. That was our first era. Circa 1990, not long after Proposition 103 had sort of ballooned in California. I think we enter a new era. We started to become curious and responsive to customers who wanted to shop via 1-800 numbers, whether it was in retailing type activities or financial services activities, and GEICO certainly had shown that that was possible. We actually built out call centers, and we easily morphed into the internet and mobile era as that clearly took over for the most part.
Just the channel alone gave us a great deal of growth, as you can see. Probably the defining element of this era was building a brand. Even though I said it was at first a little elusive and somewhat interesting, it has become an asset for us and an asset that we believe we can leverage, not just in insurance, but just as a consumer brand. Let's move to what I'll call the third era here. Again, no clarity to exactly when an era starts or doesn't start. Reminds me a little bit of some economists when they say, "Are you in an economic recovery or depression?" It's kind of like, I don't know. I'll tell you later. I think this started around 2007 when we started really doing our Progressive Home Advantage and bundling home and auto.
There is nothing novel about bundling home and auto or providing home and auto. We know that. It was a fundamental shift for us. I'd like to get you to think about the product that we're offering is bundling. Not so much the actual product of homeowners, but the notion of bundling. I'm going to use bundling as an extension of the notion of building on a relationship that is fundamentally formed from the auto insurance relationship and finding ways not to have our customers have the need to develop other relationships. That's what I mean by bundling. We're going to spend the next hour talking about the kinds of customers who will be in starring roles in this, what I'll call destination era. Who are the customers? What gives us the right to think we can do and build in this sector?
This notion of bundling, I would ask you to sort of recognize that we're not talking about moving away from anything that we've done before. The prior eras, the customer sets we serve, they are all serving as foundations. We don't intend to serve our non-standard customers any less or any less intensely than we do today. This is all additive. The notion of bundling is to provide reasons to stay with us, regardless of what that might mean, and that will likely mean we will distribute or provide other people's products to make sure our customers stay with us. What's in it for us? It's policy life extension. It is the primary currency of everything we're going to talk about today. We are going to find ways to thicken the tail. It's all about thickening the tail.
Now, a little bit of a memory test for some of you, and for those who aren't as comfortable with it, we will do a quick refresher. Last year, I introduced the notion of a model, and it started off with a map of the U.S. and then conveniently morphed into a nondescript 51-market to 1 market, and then all the way down to 1 market with, frankly, 4 customer sets. A little simplistic relative to what we use to run the business, but simple models, if well-constructed, are able to convey a lot of information. We're going to go back to that model today, and we're going to talk a lot about 1 particular sector. In fact, we're going to take a little bit of a vacation from talking about the customers that we serve very well.
I'm graying out what. We'll refresh your memory if necessary. Sam, Diane, and the Wrights, three customer segments that we over-index relative to our market share. We're going to focus simply on the Robinsons. The Robinsons. You'll learn more about them in just a second. Not who the Robinsons are today, but who the Robinsons will be over the next several years and decades. Those that are digitally responsive, how we've positioned the company, how we will position the company to serve those. These are the customers that we want to describe to you and give you by a set of our own actions, a reason to believe that we can actually capture a significant share of this going forward market. I'm going to try to do that by taking a look at some of the pieces that go to make up a business plan.
This is a business plan that will be obviously very high level, appropriate for a discussion like this. Try to put them into some sense of order. Sort of what are the assets that we can build on? What are the skills that we need to maintain, possibly improve, or even build? What other parts of the company do we have requirements where we need to build on, perhaps some capital investments or perhaps some talent investments? I'm going to leave off now and come back to this exact slide later and see if I can summarize what I think we may have heard during the next hour or so. I'm going to ask Tricia, Dan, and John to take us through their thoughts on this business case of how we can sort of graduate and acquire and retain the Robinsons of the future.
I've asked them to quantify where possible LEP. The quantification is very much to give you a sense of whether the dog hunts versus a pro forma performance statement. We'll end the day with a little bit of a lightning round. I'm going to ask Brian Domeck to come up and talk about retention, very pertinent for the day, and sort of extending our vocabulary, if you like, and our thoughts around measures of retention. We've been very happy with our policy life extension or policy life, PLE, for 12 months. He'll give you a way of thinking about that in a slightly expanded definition, and a definition that we will use going forward in sessions, whether in my writings or in our conference calls or whatever it might be. We'll expand our notion of retention. Very appropriate for today. I'll ask Why am I drawing a blank?
Steve Broz, thank you. It's not like I don't know. Steve Broz to come up right after that and talk a little bit about advanced analytics, some of which I've already mentioned. He'll choose Snapshot as an example and try to give you a little peek behind the scenes of how we can actually not only have access to big data, but tools that we can use and how we can match that with some external data. We're not going to use things that we're actually very determined on bringing to the marketplace. We have competitive reasons to perhaps keep that to ourselves, but I think you'll get a sense of what it means to be in a world of big data.
Sanjay Vyas will come up and just keep us grounded, because bottom line, it's what's happening on the market front in every state and give us a quick overview of four major states for us. Just what's happening, what are some of the key issues, and what are some of the actions that we'll do. Then we'll round that day out with John Barbagallo talking about commercial. He'll use a model not too dissimilar to the blob and some market segments, a couple of places where he's quite challenged and quite optimistic, and give us a sense of the commercial market as he sees it. I will come back, as I said, and we'll talk just a little bit more about those pieces of the model. Clearly we won't have covered everything, so we should have plenty of time for your questions and answers.
I'll ask John to kick us off now.
Good afternoon. Welcome. As Glenn noted, I along with Dan Witalec and Tricia Griffith, am going to talk a lot about the Robinsons. The Robinsons are a persona that we created to represent the segment that is the largest portion of the U.S. auto insurance marketplace, that being customers or households that bundle their auto with their home insurance. Glenn briefly shared with you the four segments, we're going to employ those four segments as we do that. Let's do a quick review of the four segments. Sam, over on the left there, we use as our base. We're showing you relative LEP, lifetime earned premium, across the bottom here. Sam is the core non-standard customer. Sam is who we were insuring back there in the early 90s, as Glenn pointed out.
Diane is a renter, Diane keeps her insurance in force consistently, her LEP relativity consequently is one and a half to two. The Wrights are homeowners, they don't bundle their auto and home insurance together. Their LEP relativity, two to three, obviously there on the right, the Robinsons. We share a three to four relativity for the Robinsons, we also add a plus on the end of that. We do that because we know the LEP relativity for the Robinsons will be much longer with us, we're confident it is longer with our competitors as well. By definition, we've been at this, as Glenn noted, six-ish years. The lifetime of a lot of these customers with their insurance company is decades, obviously our LEP will extend as we work with the Robinsons more into the future.
Where do these customers currently consume their auto insurance? Obviously, this is our personal estimates based on a lot of different data sources. As you can see, Sam over on the left there is predominantly an agency customer. Sam frequently has to pay his bill the day it is due, perhaps the day after it is due. Sam needs local representation for help. The Robinsons also skew heavily towards agents, for different reasons. The Robinsons have complex needs. Obviously, auto and home to start, probably maybe life, maybe a commercial policy there, some toys in the household, more complex needs that they need a local counselor for. You see Diane and the Wrights sort of equally balanced across the three channels, direct independent agents and captive or exclusive agents. Those customers have simpler needs.
The likelihood they can have their needs met over an 800 number. Over the internet is much higher. Slide notes, the Robinsons and agents dominate the marketplace. You can see that's very apparent. We also know that direct is continuing to grow. As we showed you last year, over recent years, actually been taking share more so from captive agents than independent agents. We obviously believe that trend will continue. As companies like Progressive on the direct side build out products those customers needs over on the right-hand side of this chart, and compelling experiences to bundle those products, we think the direct will grow a lot more on the right-hand side here. This does not at all mean that we aren't still focused on growing in our agency channel. We are. That 12% in the Robinson column there is very untapped for Progressive.
You'll hear a lot today as well about how we're going to go after that potential later on. Our share, and for those of you who aren't familiar with these types of graphs, let me make sure you're aware, the width of the bars here is the size of the segment. The vertical axis, obviously, on the last chart was the channel share. Here we're simply sharing our share by segment. Over indexed materially relative to our countrywide share with Sam, little over indexed with Diane, a little above average, with the Wrights, and then very low share with the Robinsons. We think today our share of the Robinsons is about 1%. It's not hard to look at this chart and say, "Where might the opportunity be?" We obviously do that, and that's what we're going to talk a lot about today.
I also want to note, though, that our profitability with the Robinsons is excellent. We want to grow this segment as fast as we can. A challenge, however, in growing with the Robinsons is that they don't shop much. Obviously, if their retention is good with the competition, that means they don't shop much. Survey data tells us almost 90% of them have no intention of shopping at their next renewal. More than two out of three have been with their current insurer more than five years. Think back to that 3 to 4x plus. They don't shop a lot, our challenge is growing them without attracting a ton of new business as Robinsons. Somewhat by design, but also obviously out of necessity, we are developing the Robinsons internally.
We are taking the Sams that come in our door, the Dianes, the Wrights, and developing them. Getting their brand loyalty early on in their lives, ensuring that we are there to meet their needs when those needs develop or mature. We've developed around two out of three of the Robinsons in our book today. Obviously, we have attracted as new customers, one out of three Robinsons, and when we look at those customers, they are predominantly younger Robinsons. Younger Robinsons are more inclined to shop. Although they don't shop a lot, they are more inclined to shop, but they are way more inclined to prefer Progressive. When I say prefer, that is referencing our brand tracking survey where we ask consumers, "Disregarding price, which company would you like to be insured with?" More than a two relativity here between younger Robinsons and mature Robinsons.
The new Robinsons we're bringing in are largely younger Robinsons. When we look at our customer set in aggregate relative to the U.S. population and relative to major competitors, it's pretty clear we skew very young. When you look at our current retention performance, and we know from previous slides, younger customers shop more, it's not surprising that our current retention results are not as good as our competitors. That said, if we can establish brand loyalty with these customers early on, and again, meet their needs as their needs develop, we can use that wave of customers over on the left there and develop them and raise the tide of our customer set on the right. That is the strategy. We want to be the destination for these customers.
While we are using those customers to raise the tide on the right, I assure you we are going to continue to crank the acquisition engine that is so successful for Progressive. We will continue to bring in wave after wave of Dianes and young Wrights, and use those to raise the tide on the right. The rest of our talk here is going to be around tactics around that game plan. I'm going to talk about developing Diane. Dan Witalec will come up and talk about bundling the Wrights. Excuse me. Dan will also talk about acquiring younger Robinsons. Tricia Griffith will then talk about retaining the Robinsons once they're Robinsons with Progressive. She's also going to walk through a fictitious example of one Diane starting with Progressive early on in her insurance tenure, if you will.
I think that will make our strategy a little clearer to you and make it a little more tangible. Let's start with developing Diane. Who is Diane? Diane, I would say in a word, is mobile. Diane is upwardly mobile. She is mobily inclined, technologically speaking. Diane is also very willing to be mobile with her auto insurance needs and her other developing needs if we don't meet those needs. Our strategy with Diane is to extend brand preference, to cross-sell aggressively, to target life events, and to expand our product availability. We're going to talk predominantly today about expanding property product availability, but be assured that we are working on other products as well to, again, ensure we meet all of Diane's needs as those needs develop. I would hope we're talking about some of those products with you in future years.
Extending brand preference. Today, we are in the sweet spot for Diane. The Superstore flow resonate really well with Diane. You can see stats on the left there. This is relative to the total consumer set that we survey. Way more aware of Progressive, much more likely to consider Progressive, much more likely to prefer Progressive, and more likely to shop Progressive. We want to continue to extend that brand preference. We are very active online because Diane is predominantly online. We're very active on handhelds because that's where she is increasingly as well, Facebook, mobile apps, banners, et cetera. Around 30% of Dianes today, when they shop for their auto insurance with Progressive Direct, shop on a mobile device. Once we get Diane in the door, the first thing we're working on right now, or the first thing we do is cross-sell renters insurance. Obviously, she's a renter.
Not everyone buys renters insurance. A lot do, and that is our first cross-sell offer to Diane. We've been really, really successful doing this in our direct channel, have a large book of renters business with the Home Advantage partners we work with. We'll talk later on in the program here about doing similarly with agents. The first thing we do is try to get the cross-sell offer in front of Diane. We also work hard to ensure we are in front of Diane when she goes through what we call life events. A life event, as an example, is getting married. Shopping goes up when people get engaged, when people get married, because normally it is, "Are we going to go with your insurance company or my insurance company?" We got to be there when that discussion is happening with the reasons to stay with Progressive.
When Diane moves, also much higher frequency of shopping. We're working hard to be there as well. We are the exclusive insurance advertiser with the United States Postal Service Change of Address program, be that online or in hard copy. We're not outstanding yet, I would tell you, in addressing life events, but we're getting better, and it's critical because that's when Diane shops. Let's talk a little bit more now about the cross-sell offer and renters specifically, but more broadly, expanding product availability, where, when, and how Diane wants to buy, in this case, property insurance, more specifically, renters insurance. I'm going to cross over a bit into home here as well by necessity, but predominantly, I want to talk about getting the product to Diane when she needs it.
Within our agency channel on the left there, we have Progressive Home Advantage inclusive of our homeowner offering on a limited distribution basis. We're growing that. Dan will talk more about that in a moment. What I want to tell you about today is that a week from today, we will start writing renters insurance on Progressive paper. Progressive will be the insurer. We are very comfortable with the risk profile of renters insurance, high frequency, short tail, very little catastrophe exposure there. It will be on our paper, on our balance sheet, on our P&L, but serviced by ASI. American Strategic Insurance, a company that we have a small financial interest in, will do the servicing for that renters product.
I should have already mentioned, and I will mention it again later, I hope, when we can cross-sell Diane with renters insurance, our retention of Diane goes up more than 20%. Renters premium is not going to be huge. I fully acknowledge that. The incremental auto premium is very material. Again, more than 20% when we can cross-sell Diane. Our first state for renters will be Ohio, as I said, a week from today. We'll roll out a couple of other big states this year, round out the larger states next year, and then fill out the rest of the country by sometime in 2016. We also cross-sell Diane in direct. The next three columns, if you will, are predominantly focusing on our direct customers.
When Diane calls to pay a bill or for any other service reason, we key prompts for our reps to make the cross-sell offer. Today, those offers either go to a single Home Advantage carrier or to a unit internally that we call our multi-product sales team. There as well, we're selling one carrier. For renters, that works great. For homeowners, it doesn't work optimally, and that is because homeowners are relatively select in their underwriting, I'll say, meaning that it is not uncommon for a homeowner carrier to reject even quote one out of three of the quote requests they get. Very different underwriting appetites across homeowners carriers. What we found is that by creating an in-house agency, Progressive people with access to a broader set of homeowners carriers, we have a much greater chance of being able to meet that customer's needs.
Next month, we are going to roll out a multi-carrier platform within that group. That group has sort of been in pilot mode for about a year. This system will give us the ability to scale that group, and we expect to triple, maybe quadruple that group by year-end. We expect to double it again next year because we have found from our pilot this is a much better experience for our customers. We are much more likely to be able to have a product for them and have a competitive product for them. We've been a leader when it comes to offering property on the desktop, obviously, with the help of our Home Advantage carriers, who are pretty advanced online. We have the most robust, I would tell you, multi-product quoting experience.
That means coming to progressive.com and getting an auto and home or auto and renters quote together in one experience. We also obviously offer links to opportunities to quote your home or renters when you're online servicing your policy with Progressive. We've been a leader here. Property online is somewhat in its infancy, but certainly growing, and I'm really excited to tell you that per Comscore's most recent benchmarker study for property insurance, Progressive Home Advantage is the leader for renters quotes online. As you can see, a leader by a factor of around two relative to the next competitor. I said property is in its infancy on the desktop. It's even more so in the mobile space. Remember Diane. Diane is mobile. We got to be there for Diane in mobile. We now have renters quoting on mobile devices countrywide.
The multi-product quoting experience that I shared earlier, we have in the process of rolling out. Today, we have, I think, 10 states elevated, and we expect to elevate the rest of the country in a month or so. Again, the intent here, as Glenn noted, a policy life extension. When we can cross-sell Diane more than a 20% lift in policy life expectancy. What we're finding is that is actually just the beginning. If we can establish that brand loyalty early, cross-sell the renters early, when Diane buys her first home, if she already has her renters with Home Advantage, she is more than twice as likely to convert when she quotes her home with us than if she didn't have the renters with us already. That's creating that path. That's creating the ability to use the wave to raise the tide.
Glenn noted we're going to try and give you some perspectives on the benefits of the initiatives we're talking about today. In the instances of looking at Diane going through life events, again, getting married, buying her first home, we think, based on some studies we do on defection, that we lose 35% of Dianes when they go through this life event. We're confident that we can reduce that materially, and when we estimate the benefit of doing that, we come up with a range of between $140 million and $230 million incremental LEP annually. Okay? Incremental LEP annually, it's a wide range because we are layering on numerous estimates of the different things we're doing. I'm sure incremental LEP annually is not a concept you normally think about.
It is one we do because we use this to prioritize efforts internally, and we use this because we want to build that invisible balance sheet that Glenn noted. Let me give you a simple example to hopefully make sure you understand the concept of incremental LEP annually. Let's say from the previous example, the point estimate was $200 million incremental LEP per year. Let's assume further that the PLE for the customer is five years, and let's assume as well that the benefits of these efforts are earned equally. What you can see is that the five years plays out obviously over the subsequent calendar years, and you don't ramp to the full $200 million on a calendar year basis until year five.
Again, we think it's a great way to look at things, is the way we look at our prioritization efforts internally, and you'll see the similar metrics in the course of Dan and Tricia's talks as well. Again, our strategy with Diane, extend brand preference, cross-sell aggressively, target life events, and expand product availability. We can be Diane's destination, and we can use wave after wave of Diane to continue to raise the tide of Progressive customers. I'm going to hand the floor over now to Dan Witalec, who is going to talk through bundling the Wrights and acquiring young Robinsons.
Good afternoon. John gave you an overview of how we're going to grow the Robinsons broadly, and one specific strategy we're using to grow them, developing Dianes through life events. I'm going to hit on two additional strategies to grow the Robinsons, bundling our Wrights and acquiring young Robinsons. Let's start with bundling Wrights. Let me tell you a little bit about the Wrights. The Wrights are homeowners who have their auto and home insurance with separate carriers. While they're unbundled today, we believe, and we see in our data, that this is a transient state for most of these customers. At some point in their life, they are likely to bundle their auto and home insurance together. Our challenge is to get them to bundle Progressive Home Advantage with our auto, frankly, before a competitor lures them away.
That's really what we're talking about when we talk about the Wrights. It's a retention play to get them to bundle Progressive Home Advantage and stay with us longer. The good news about the Wrights, as John showed, is that we have high share with them today. We acquire a lot of Wrights, and obviously, you can't retain somebody that you don't get in the first place. That's certainly a key first step in our strategy. How are we going to cross-sell Wrights, and get them to bundle Progressive Home Advantage? I'm going to talk about three specific tactics to do that. Homeowner capacity, agent acceptance, and marketing to existing customers. Let me start with homeowner capacity. I think you're all aware, for our Progressive Home Advantage product, we do not underwrite the homeowners product ourselves.
This is distinct from what John talked about in terms of our agency renters product that we'll be rolling out soon. We do have a group of 11 PHA companies who underwrite the insurance with us. It's a large group of companies. Together, they represent about 6% of the total homeowner market. You can see from the middle graph here that in general, they've been profitable. They've been more profitable than the overall homeowner industry over the last five years. Importantly, they're financially stable. If you take an average of those 11 companies, they have a 1.4 premium to surplus ratio. A little bit higher than the industry average, but still connotes financial stability. Importantly, they have the capacity in their homeowners product to grow as we grow with them. The Wrights have a strong share in the independent agent channel.
Obviously, if we want to cross-sell Progressive Home Advantage to them in the agent channel, we have to have a product available for our agents. ASI is our key underwriter in the agent channel. You can see in this chart our rollout plan by state for our Progressive Home Advantage product in the agency channel. Through 2012, we had rolled out PHA and agency in states that represent 52% of the countrywide premium. We added Florida in 2013, and we have plans for an additional seven states in 2014. If all goes to plan, by the end of 2014, we will be in states that cover 84% of countrywide premium. We're not in every agency in each of these states, an additional tactic that we're planning is to roll out to additional agents. In fact, over the next year, we'll double the number of agents who have Progressive Home Advantage in their office. Between adding new states and adding agents in those states, we have a real opportunity to increase our distribution in the agency channel.
We're talking about agents. Obviously, another critical component is we need agents to think of us as a preferred company to put their best customers with us. Our heritage in the independent agent channel is as a non-standard competitor, but we've really changed agents' view of us over time. Two data points to support that. First, this graph shows that in 2013 versus 2012, 6% more agents said they would put their preferred customers with Progressive. That's one market research data point. We certainly like the direction of that.
Here's some behavioral data that shows that agents, over time, there's a higher quote mix of Wrights and Robinsons. Think of preferred customers that agents are putting with us. Two encouraging points that agents think of us more as a preferred company. The last topic I want to hit on in terms of cross-selling Wrights, and making them Robinsons, is marketing to existing customers. Glenn referenced early on that we're proud of the progress we've made in our broad market advertising, largely through Flo and the Superstore campaign. What we still need to work on a little bit, and we have more opportunity, is in the related but distinct skill of marketing to existing customers. The first challenge in marketing to existing customers is simply to make sure that they know that we have a Progressive Home Advantage product available.
This chart shows the awareness among our existing Progressive customers of our Progressive Home Advantage product. You can see over the last five years, we've increased awareness among our existing customers from about 40% to a little over 60%, largely driven by some mass market ads that we've had out there. Each of those different screenshots is a different TV bundled ad that we've had. We've made a lot of progress here, but at about 60% awareness among our existing customers, we think there's a lot more potential here. To reinforce the importance of awareness of our existing products, here's some data that shows the link between awareness of our products and consideration for Progressive. You can see on the left that if a customer is not aware of Progressive Home Advantage, nor Snapshot, we'll throw some Snapshot data in here, they have consideration Progressive about 13%.
That increases significantly when they're aware of either Progressive Home Advantage or Snapshot, and more than doubles to 28% when they are aware of both Progressive Home Advantage and Snapshot. Just an additional data point, simply making our customers aware of the products that we have is really important to their consideration for Progressive. How do we do it? How do we make customers aware of our home product? First things first, we have an existing relationship with these customers. We want to make sure when they are servicing their policy, we let them know about Progressive Home Advantage. This slide shows that if a customer comes online to service their auto or motorcycle policy, we make sure that we point out Progressive Home Advantage, and we're doing this in the call center as well.
We're also leveraging some one-to-one media, like email, to let our existing customers know that we have Progressive Home Advantage. It's not just blasting our customers with a ton of emails or snail mails. It's really being smart about how we market to those existing customers. On this chart, on this slide, I show an example of direct mail marketing to our existing auto customers where we're trying to sell them Progressive Home Advantage. We have found that if we market to customers around their homeowner expiration date, you can see in the lower left-hand chart there, customers respond to that direct mail piece at a rate greater than 3x when we don't use homeowner expiration date. A huge increase in response rate with similar conversion. That means that our cost per homeowner sale when we use homeowner expiration date is about 70% less.
Admittedly, we're not the first to use homeowner expiration date, but it's, I think, a good example of where we're going in terms of leveraging more data to market more efficiently and more smartly to our existing customers. Where are we in terms of our multi-product agenda? How much progress have we made? On these charts, I show the percentage of auto customers who have more than one product with Progressive. Direct's on the left, agency's on the right. Lot of colors here, let me just explain real quick what they are. The light blue on the bottom is special lines, think motorcycle, boat, RV. The orange is Progressive Home Advantage, both homeowners and renters. The white is Umbrella, the dark blue are customers who have three or more products with Progressive.
You can see that we've grown pretty dramatically across the board, but especially in direct through Progressive Home Advantage. In fact, we now have over 1 million Progressive Home Advantage policyholders. Took us about 7 years to get those 1 million. As a reference point, it took us over 50 years to get 1 million auto policyholders. You'll also notice on this chart, agency has not grown as quickly in terms of multi-product customers as direct has. I think the encouraging data point here is that if you look at the special lines and our Umbrella multi-product rate in agency, it's as good or better than as in direct.
We think we have real opportunity with Progressive Home Advantage in agency as we roll it out to more agents and more states, as I showed earlier, and we roll out our renters product broadly, as John referenced earlier. We have a lot of potential here to increase our Progressive Home Advantage penetration in agency and ultimately increase retention there. What is this all worth? Again, what we're talking about here in bundling Wrights, getting that unbundled customer to bundle home insurance with us and become a Robinson, it's all about retention. This decay curve shows the improvement in retention. We think it's about a greater than 20% improvement in policy life expectancy if we can get that customer to bundle.
We think that's worth about $90 to $170 million in incremental LEP per year if we're able to continue our growth rate in terms of penetrating multi-product customers. That is a big opportunity for us as we think about growing Robinsons. That's bundling Wrights. Let me now turn to acquiring young Robinsons. John laid out earlier that realistically, we are focused on acquiring young Robinsons as opposed to Robinsons more broadly. They shop more and frankly, they like us more. We think largely that's driven by the fact that a lot of these young Robinsons came of age during the retail era that Glenn referenced earlier. Let's talk a little bit about the young Robinsons. These are affluent young families who have growing insurance needs. Importantly, they are very technology savvy, especially relative to older Robinsons.
These customers use digital media a lot and really have a high expectation about how they want to interact with companies that they use. How are we going to acquire these young Robinsons? Let's talk about three specific tactics. Digital target marketing, online homeowner quoting, and some compelling product features. We'll start with digital target marketing. I mentioned that the young Robinsons are technology savvy, so it makes sense that we're going to target them via digital means. The nice thing about digital marketing is that it can be more targeted than traditional mass media, like a TV ad.
For example, most of our Robinsons come to us via online, either desktop or mobile, we're able to see the types of sites that they visit, the browsers that they've used, the devices that they've used to quote us, and we're able to build some lookalike models based on that data to predict other consumers who might also be Robinsons. What that ultimately means, you'll see in the chart on the right, is that we're able to double the percentage of quotes who are Robinsons by using these targeted digital ads as opposed to more general online ads. The other nice benefit of this is that we're able to send them a specific ad by targeting them.
While our general messaging works well with the young Robinsons, we're also testing some unique messages to them, like bundled comparison rates, which I'll talk about in a second, and loyalty ads, which we think can really resonate particularly well with this segment. All right. Now let me turn to the second tactic in acquiring young Robinsons, online homeowner quoting. John shared with you a graph from Comscore that showed our share of renters quotes on the desktop where we were the leader. This is a similar graph, but it shows the share of desktop homeowner quotes. You can see on here that by more than a two to one factor, we are the leader in desktop homeowner quotes. Let me stop for a second on that. I told you home is still a new product to us.
We don't even underwrite it ourselves, and yet we are the clear leader right now in online homeowner quoting. Admittedly, homeowner quoting is very small relative to auto quoting, but we think it's only going to grow. This fits what we do really well, and we think we're in a really good position here on online homeowner quoting. Part of how we're leading in online homeowner quoting is through multi-product quoting. John referenced this earlier. This is simply entering your information one time and being able to get both an auto and a homeowner quote at the same time. Really what we're doing is leveraging the leadership we have in auto online quoting and pulling it into homeowner quoting as well. We think another natural extension of that multi-product quoting is bundled comparison rates. Let me take a step back.
I think you're all aware if you get a quote on progressive.com for auto today, at the end of the quote, you can get comparison rates of some of our top competitors. This has been a really great feature for us. It's a feature that we advertise heavily and has worked very well in building our brand and acquiring customers. We think a natural extension of that, as we now have bundled rates, is to have bundled comparison rates at the end of that quote as well. We think this can be a really compelling message to these young Robinsons to get them to come to Progressive and all in one fell swoop, get a bundled quote and compare competitor rates. Of course, it doesn't matter what we think. We're testing this in the marketplace now. It really matters what consumers think.
Here's a TV ad that we're using as we test whether or not consumers like this feature of bundled comparison rates.
We compare Progressive direct rates with other top companies so you get a great price. Now you can compare your Progressive direct auto rate bundled with home insurance too.
Okay. What does this remote do?
You know, comparing bundled home and auto rates too. That's Progressive.
All right. I'm going to finish up on acquiring young Robinsons by talking about two compelling product features. First, let me hit on Snapshot. We haven't talked a lot about Snapshot today, but know that it is still an absolutely critical product for us. It is now over $2 billion in premium for us. In fact, year to date, our premium with Snapshot has grown about 30%. It's a critical segmentation tool for us, probably most important for the story that I'm telling today, Snapshot resonates really well with Robinsons and young Robinsons in particular. 40% of new Robinsons customers choose Snapshot during the quote, versus about 34% of other customers. We also know that young Robinsons in particular are particularly comfortable with Snapshot technology. I can't talk about Snapshot without giving you some data on the retention benefits that we've seen from Snapshot.
Among all customers who enroll in Snapshot, we see an 11% greater policy life expectancy. Among those customers who get a discount through Snapshot, it's a 19% higher policy life expectancy. This is a really great tool for us, both to acquire new customers and to retain the customers that we have. Let me finish up on product features by talking a little bit about Umbrella. Umbrella covers you in excess of your auto and homeowner insurance limits. It is admittedly a small product for us and for the industry. For the Robinsons and young Robinsons in particular, we think it's a critical product. For example, Umbrella customers are more than three times more likely to be a young Robinson than non-Umbrella customers. From a retention perspective, we see that Umbrella customers have a 34% longer policy life expectancy than non-Umbrella high-limit auto customers.
Again, if we were thinking about the Sams of the world or we didn't have a target, Umbrella probably wouldn't be at the top of our list. As we are focused more specifically on Robinsons and young Robinsons, Umbrella's a great tool both to acquire those customers and to retain them. What is all this worth acquiring young Robinsons? First, I should say that we've made progress over the last several years in increasing the number of new policies who are young Robinsons. You can see some of that growth on this chart. It dipped a little bit in 2012 as we raised rates across the board, but has gone up over time. By staying on this trajectory or increasing it, we think there's a $45 million-$100 million year opportunity in incremental LEP. Again, using that same incremental LEP that we explained before.
This is the smallest of the four opportunities that John, Tricia, and I will talk about today, but a lot of the things that we've talked about here, like Snapshot on Umbrella, are great for acquiring young Robinsons, but they have an additional benefit in retaining as well that goes beyond the quantification that I've shown here. With that, I'll turn it over to Tricia to talk about retention.
I'm going to wrap up our section about what we're doing to give the Robinsons a reason to stay. The retention component of our business plan is a critical part of our success, and I'm pretty excited about the things we're working on and things we're going to continue to work on throughout the years. Before I do that, let me talk a little bit about the demographics of the Robinsons. I'm not going to go into similar to the young Robinsons, but I want to draw your attention to a couple of things. One, the Robinsons are increasingly. Every once in a while, they want that personal touch. Well, it's a balance. Good service, expect convenience. We need to think about that as well.
In fact, there's a lot of companies that they'll benchmark us against, and we need to make sure we exceed those expectations. Not just insurance companies think about our evolving book of business. Our strategy today, to retain the Robinsons is going to be about responsive products and services, our continued investment into mobile. Is it working? Sorry. I think I was talking loud enough, wasn't I? Our continued investment into mobile, our low cost structure, and our cares for customer marketing. We're going to talk a little bit about that. Then I'll go into what John talked about in terms of going through an evolving Diane, one person, kind of her whole insurance journey. What you're seeing here is not new news to you. I've showed this over the couple of years.
What you'll see on the left-hand chart, on the y-axis, you'll see policy life expectancy. With that, you'll see all auto customers and auto customers who own a home. As you'll see, when we have more products in a household, it's stickier. We retain longer, and in fact, even longer when there's an auto and a home. We know this, it's important to us. In addition, I'll talk about NPS, Net Promoter Score, for those of you that recall. That's really our customer service measure that we use throughout the company. In addition, using the same cohort segment of customers, the longer they stay, the happier they are. The reason I continue to share this with you is because we've been testing all the things we've talked about earlier today for the last seven years. We know what we're doing is working.
That's really the important part of this. In fact, we have more than doubled in the last five years, customers that have more than one product. This is working. We'll continue to look at this as we develop more and more ancillary products for our customers. I talked a little bit about the Robinsons becoming more and more digital. They appreciate the transactions. We try to make the routine things for them easy. However, they will occasionally want something more. We created a preferred customer call center staff that can take these calls. Think of it when they want to cross-sell, and it's a little bit confusing, or we want to be able to tell them about the loyalty that they've earned.
What we have seen through this pilot is that when people call in to this preferred customer set, the NPS is lifted by five points. We know this is working. We know this is low in frequency, but also very important to them. That balance of being able to have all the digital things that they need, plus the occasional touch is really important. What you see here is payments by three different shares. Obviously, a big delta between what you see Diane doing and what you see the Robinsons doing. She's making a lot of her payments on a mobile device. You might say, "How are you growing for the Robinsons when they're such a low percentage of making payments?" That's exactly the point. Our Dianes of today are the Robinsons of tomorrow.
There's a Diane today that's going to be a Robinson tomorrow, a month from now, three months from now. We're already set for the future Robinsons. We'll continue to see what the needs of Diane and the Wrights are, and we'll grow that for the Robinsons. They're perfectly positioned to be able to make their payments when that Diane becomes a Robinson. In fact, it wouldn't be rational for us to think that the moment Diane bundles her home and auto with us, she's going to start writing via checks. This tells us that what we've been working towards, we're set to, and we won't stop here. We'll continue. Clearly, the young Robinsons fall more in line with the Dianes. Again, we're set there. I talked about investing in mobile. There's a couple things that we've recently done.
What you're seeing here is being able to have all of the items, all of the insurance products that our bundlers have in one place, even on your mobile phone. They'll be able to do anything they can, anything they need to with their whole portfolio of insurance needs. This is pretty important. The next thing is near and dear to my heart, and we just rolled out a new loss reporting app. Now we've had the ability to report losses online for our customers for many, many years. This is a very slick app. You can add pictures, it's quick, it's efficient, and people are going to demand it because they bought online, they pay online, and they want to report their loss online.
When I tell you about this, whether it's looking at the bundlers' insurance needs all in one place, or being able to report a loss online, we really see this as the table stakes. This is the cost of admission. There's a lot of advertising out there that says a lot of slick apps, we really have to be doing this. It's a requirement. This is investment we'll continue to make based on the fact that our customers have told us they need it. For several years now, I've talked about our service centers and how we believe it is the best place if you ever do have an accident. Our goal has been to have significantly more customers choose the service center and have experiences unmatched in the industry.
The first thing we needed to do was to make sure that our employees, our current customers, our independent agents, were aware of our service centers and were aware of what we were doing at our service centers. We set out to relaunch our service centers a few years ago, and clearly it worked because the usage is up. We look at this at cars per day. Our usage is up in the past two years, over 60%. Now, even if you add in some of the new service centers that we've added, it's still up over 40%. Significant usage based on what we believe is more awareness. We're going to continue to work on this, and we believe as more and more people experience this, the barrier to acceptance really diminishes.
The great news about this is that Robinsons already use the service center more often, and they like it better. They have the higher NPS. We've had over a decade of experience with Robinsons because when our insurers are involved in an accident with a claimant, oftentimes those claimants are Robinsons from other companies. They've been able to give us insight for the last decade, how to really design the process of the service center and what really makes us be able to exceed their expectations. This is really key. We've learned a lot over the past decade because we've had that experience, even though our market share currently is low. Our advertising is absolutely helping too.
We do something we call a defection survey, and we asked our insurers, "What motivated you to shop for an insurance quote?" As you can see, far back as 2008, oftentimes their answer was, "I want to buy my insurance from one company, all my products from one company." Of course, through our advertising and through being able to give people the reason to stay and have products and services they need so they don't need to shop, that has clearly diminished. We still have a lot of work to do, and that's what this whole transformation is about, making sure as we go into this destination era, that people know we'll be able to meet all their needs with Progressive, and we'll continue to learn from our consumers as this continues. Glenn talked earlier today, and Brian talked last year about cost.
The cost structure is clearly very important. You have to have a competitive cost structure. What you are seeing on both of these charts are five-year combined annual growth rate of direct written premium. The left chart is compared to our expense ratio, the right chart to our loss expense ratio. These are the top 10 companies, clearly the message here is the companies that have a low competitive, I should say, cost structure are the winning companies in terms of growth and profitability. This is really important. We are going to invest a lot, but you have to remain competitive. We talk about this all the time, this is something that we have worked hard at. We are well-positioned, we are not going to give up that position. The even better news is that the Robinsons have a lower expense ratio.
Normally, when we talk about expense ratio to this group, we will look at calendar year expense ratio, we will talk about it in the aggregate. What this chart shows you is lifetime direct expense ratio with the segments that we specifically talked about today. As you can see, the Robinsons are lower. I think the headline here is we can play in the preferred space. Now, you might say, are the Sams subsidizing the Robinsons? No, we actually look at loss ratio compared to each segment and make sure that we do the right thing with each segment in terms of pricing. We know that as we get more Robinsons on the book, that our overall expense ratio will decline. This just tells us we know we can play in this preferred space.
Last year, I talked about the customer perception, we talked about the balancing of value and quality, we have always been really well-positioned from a value perspective. Think of it in terms of good cost for what you get and savings. So that we feel good about our position there. From a quality perspective, we are happy with it, we know we can do better, we talk so often about how we can get better in quality internally that we are disappointed that it has not moved more. Think of it in terms of honest, open, caring for customers, leadership brands. Some of the things are a little bit less tangible to get your arms around.
This year, or actually last year, we stepped back and said, "How can we talk to our customers and have them feel the way we feel about them?" Think about things in terms of culture, what Glenn talked about earlier, core values, the things that we believe in every day. We stepped back and said, "Well, how can we do sort of a campaign about this?" I hesitate to say a campaign because it is more of an overlay of Flo in the Superstore. We asked the question, what type of company would Flo work for? As we started to build answers to that in terms of never finishing until our work is done, always trying to make things better, we really saw it come together.
In fact, we were able to link it to Flo's apron, and we said, "What does an apron have to do with car insurance?" I'm going to show you an ad we recently rolled out. This is one you will not hear on prime time. This is something, you have the voiceovers, and one makes it, and several others hit the floor of the marketing room. This one did not hit the cutting room floor, actually. This one hit the cutting room floor. I will tell you, the 26,000 Progressive employees all voted for this one. When you hear the voiceover, hopefully you recognize it. We certainly like it, and now I'll show it to you.
What does an apron have to do with car insurance? An apron is hard work. An apron is loving what you do. An apron is not quitting until you've made something a little better. What does an apron have to do with car insurance? For us, everything.
Did you guys recognize the voice? Please tell me. I want a job after today. Actually, this ad internally has really inspired all of us to figuratively wear the apron every day for the customers we're so privileged to serve, for our shareholders, and for each other. It's really been a great thing internally, and we'll continue to share it externally to get some movement and show our current customers and future customers how we feel about quality and what we can add to that. To wrap up our decay curves, this is the value of retaining Robinsons. We see the worth of this somewhere between $190 million and $390 million incremental LEP per year. Remember, when we're showing these decay curves and the incremental LEP, this is only one part of our business plan. Think of it in terms of just the Robinson plan.
We have other independent growth plans that we're working on. This to us is really big. Obviously, we want less and less Robinsons to defect. If we can do the things that I talked about today and other things that we think about every day and design every day, we believe that there's a huge value in this. In fact, overall, when you combine what John and Dan and I all talked about, whether it's graduating Wrights, bundling Wrights, acquiring young Robinsons, we believe that the incremental LEP per year value is just under $900 million. We think it's really significant, and we're really excited to be able to continue to move towards this destination era and do the things that are on our playbook. I'm going to step back, and I'm going to talk about Diane.
I'm going to walk you through a journey of Diane's life. Remember, there's lots of combinations of Diane. Don't think that this is solely the one we're working on. We're thinking about lots of different combinations. In addition, you'll see some PLE and LEP numbers. They are best estimates, but obviously, they can't be exact because there are different combinations. Let me give you a little reminder. Diane is direct biased. She is an auto customer. She rents, but doesn't necessarily have the need for renters insurance. Below, her average PLE is about 40 months, and her loss is about $4,600. Assume her average written premium is just shy of $700 for a six-month policy. We're going to start a relationship with Diane because she chose us. What do we know about Diane?
We know that she's digitally savvy, she's a perfect candidate for Snapshot. She puts her Snapshot device in her car. Lo and behold, she gets a discount, and her PLEs extend. What else do we know about Diane? Well, she turned 30 this year, and she grew up on a lake and has always wanted to boat, she has been saving up for the boat of her dreams for the past three or four years. Her parents decide to give her a sum of money to wrap up that purchase, and she's able to purchase a boat for her 30th birthday. She gets the boat. Now she's a special lines customer. We also know that she gets the boat with premium coverage. We call it Propulsion Plus internally. This was something that our customers told us they wanted.
Several years ago in claims, we would hear about, we want something more than if just an accident happens on the water. We'd like something that could help us with almost warranty maintenance related things. We added this additional coverage, and people are willing to pay a premium for it, and she was able to do that. What do we know is she, on the weekends, has friends on the boat, and those friends bring friends, and one weekend, they brought a young man named James, and they became smitten with each other. Next thing you know, they get engaged. Good idea to buy the boat. Anyway, they get engaged, and for now, she actually has a reason to have coverage. She wants rental coverage because she wants to insure that diamond that James just gave her.
She easily goes on to our PHA online, gets a rental coverage for her ring, a rider for her ring, and now she's a PHA partner with us. Very significant. What do we know now? We know that they get married the next year. This is a pivotal time. John talked about this time in terms of events. This is the time where we could lose James or lose both of James and her to another company or get James on Progressive policy. We've been following her. We know she's engaged, we know this is hopefully upcoming. We're able to internally market to her about having him come on our policy, and he does. Now we have two autos. We could add wedding insurance. We don't have it now. Would we build a big claims organization around wedding insurance? No.
If we find that our customers have demands that we can distribute and it makes sense, we'll do that. That's really the important part. This is the slide where we know that this is a pivotal time for the event, and think about all these as events that we know now, but a lot of them know actually predictably, and that's really where you want to have that relationship continue. What do we know next? We know that they're expecting a baby. They move out of their apartment and buy their first home. This is a little bit more of a complex transaction. They call into our in-house agency, and she talks about who's the right partner for me with my home, and we're able to set her up with that right partner.
This is a significant event, and effectively now they have become Robinsons. Additionally, in that conversation, they talked about a couple other needs that they might have, and she just said, "You know what? We'll put things aside, but I'll think about them in the future." What do we know now? We know that one of the things that they talked about was a Deductible Savings Bank. She started participating in that, and they have their baby, Rachel. What do we know next? We know that James, we don't know if it's a midlife crisis or what he's doing, but he's looking for a classic car. He calls in, and we give him a quote for a classic car. They decide not to get it, but we know a little bit more about James, so we're putting that away, right?
We do know that they buy MBI, which is the mechanical breakdown insurance. Think of that as a warranty insurance, and it's a product that we'll have available later this year. They don't end up buying the classic car, but they do end up buying a puppy for Rachel. We're able to offer them pet insurance. Unfortunately, the next year, they have a little fender bender, but fortunately, we have a service center nearby. They're able to use the service center. We know when people use the service center, they have great experiences, and they stay with us longer. To summarize a decade of Diane's insurance journey, you'll see we've been able to offer a significant amount of products and services when she has needed it.
Now as we summarize the decade, now it's not just Diane, it's Diane, it's James, and in the future, Rachel. We have a decade of needs, and you can watch, and the PLEs are the orange dot and the blue chart of the left. As you can see, we've more than doubled in a very short amount of time. We don't intend to stop in 2024. We want to have another decade with Diane and her family and be able to have the products and services that give her a reason to stay. I'm going to wrap up our section. John talked about waves and tides, and our intentions are this, to continue to be an acquisition machine, an acquisition engine, to continue to acquire those customers.
We know those Dianes and Wrights and young Robinsons, as they evolve in their insurance journey, we're going to give them a reason to be a Robinson, to stay with us based on the needs we know of the Robinsons. This $5 billion opportunity that you're looking at here, this just gets us to the average share we have now, the average market share. We know there's even a lot more upside. We have a lot of work to do to get here. We're very well positioned. We've shared with you just a few chapters of our story today, and we very much look forward to sharing additional chapters with you as we continue with our business plan. Now I would like to introduce Brian Domeck, who will start our lightning round.
Today, we have talked a lot about extending policy life expectancy and increasing lifetime earned premium. For the next few minutes, I will give a brief review of how we calculate PLEs, and then give a preview into a few additional retention measures that we plan to incorporate into future external releases, measures that I think will enhance our disclosures. Policy life expectancy, in simplest terms, is the average length of time that we expect a policy to stay with us, measured in months and years. In mathematical terms, it's the area under the curve. It's a measure that, for us, has many applications in the business. We use it to help forecast policies in force, and that helps us inform and aid staffing decisions, particularly in claims and customer service.
It's also a key input into determining our allowable acquisition costs, and therefore, what we choose to invest in advertising and what we choose to spend on advertising, both of those critical investment decisions. Calculating PLEs involves estimating a series of month-to-month retention ratios. For example, the one-month retention ratio is the number of policies or percent of policies we expect to stay with us one month after they incepted. For the Robinsons, it's pretty close to 100%. For some of the other segments, a little bit less. We have many, many months and years of data to make those estimations. For purposes of calculating PLEs, we use the trailing 12 monthly measures. We use trailing 12 as a measurement for several reasons. One, it's more indicative of recent experience. Two, it provides a relatively stable value, and it also addresses seasonality, as we've seen from our own experience that the PLEs actually differ based upon when a policy incepts.
Similarly, we'll calculate a two-month retention ratio. What percent of policies are with us after two months? We'll continue on for subsequent months. Here, I've indicated through month seven. I picked month seven because it is the first renewal event for our policies, which have a six-month policy term. For many people, it's a critical decision. It's often the first time they may see a rate that differs from when they incepted with us. It tends to be a critical decision point, a little bit more so for the Sams and Dianes than the Robinsons. These monthly retention ratios yield the PLE decay curve.
For now, we use 60 months of historical data to project PLEs and then extrapolate out to month 300, 25 years. I say for now because as we get more of our policyholders to stay with us 10 years and 15 years, as Glenn referred to it, thickening that tail, we'll have more historical experience to be able to project those projections beyond 60 months. The aggregation of those monthly retention ratios yields that area under the curve or the PLE. How are we doing? In general, the trend is upward. We had a little bit of setback in 2012 and 2013, in that period of time shortly after we raised the fair amount in the second half of 2012. Over this time period, PLEs are up about 20%.
Hopefully, one of the key messages you hear today and take away from today is we believe there's much upside to this measure. As we are successful in many of the opportunities and tactics that John, Dan, and Tricia referred to, I think we'll reach much of that upside. For external reporting and what we report. Historically, we have reported this year-over-year change in PLE, both for agency auto and direct auto. We do it by channel. Last year, in the second and third quarters, we were reporting PLEs were down in the neighborhood of 6%-8% in both channels. In the first quarter 2014, which we issued on Monday, we reported agency was down 2% on a year-over-year basis, and pleased to say that direct PLEs are now up 3% on a year-over-year basis.
I've mentioned some of the benefits or reasons we use the trailing 12 measure, but it is still a lagging measure. In fact, last year, even while reporting PLEs being down, we saw other measures, more leading indicators that our retention rates were improving. It's those more leading measures that we want to incorporate into future external reporting. What's the first one? The first is just calculating PLEs on a trailing three-month measure. It's obviously more responsive to current experience. It doesn't address seasonality real well, but on a year-over-year basis, it can be very comparable measure. Overlaying this trailing three-month measure, you can see it does create a little bit more volatility, but we have reported PLEs decreasing earlier in the second half of 2012, shortly after we raised rates. Certainly, we have been reporting PLEs increasing faster.
In fact, about five to six months earlier than a trailing 12-month measure. I think both these measures have their merits. We plan to incorporate both in terms of our commentary and disclosures in future quarterly reports. Another measure that we plan to discuss in our quarterly reports is what we'll call a renewal ratio. Think of that, of all the policies that have come up for renewal during a period of time, generally a quarter, what percentage do renew? Think of this as the aggregation of that month 7 event I talked about earlier, also the month 13 renewal event, the month 19 renewal event, et cetera. The aggregation of all those renewal events. Again, this is more of a leading measure because last year we saw this measure improving as well, even as we were reporting the trailing 12 PLEs down.
These leading indicators, leading measures, are what we want to include in future external reporting. You'll see them starting in the second quarter Q. The final measure, which is going to be more of an internal measure for us, is the growth in policies that have stayed with us more than 10 years. We're thinking now of PLEs, not just in terms of months and years, but now in terms of decades. As we are successful in doing the many things in terms of attracting and developing and retaining Robinsons, I'm confident that this measure, the measure of the growth and absolute counts of policies in force that have stayed with us 10 years or longer, the thickening of the tail, will continue to improve in future quarters. With that, I will now turn it over to Steve Broz.
Analysis has long been a strength of Progressive. It's one we've leveraged in every era to be successful. What we're excited about is the ability to take this long-held strength and combine it with new tools and technologies, they're often called big data tools and technologies, in a way that allows us to maintain and extend our competitive advantage. We could talk about this in any number of areas inside the company, but time is limited today, so we'll talk about it in pricing segmentation. This is a graph that Glenn shared early in the presentation. As a reminder, the orange line is average pricing, where everyone pays the same price. The red line is the theoretical limit that we think of as perfect pricing. What you see in between is 20 years of continual improvement of our personal auto product.
It includes the yellow line there is eight three, a product we expect to release later this year. The two things I'd like you to take away from this slide. One is that over 20 years, the annual rate of improvement in this measure has been maintained without almost any change over 20 years. The other is that as good as we are at this, there's plenty of room to improve. We'll continue to invest in pricing segmentation. One of the investments we've made is in Snapshot, and you can see that the data we get from vehicles that have a Snapshot device installed improves segmentation even further and continue to get us closer to perfect pricing. Remember, that's not the only thing we get from Snapshot. As Dan pointed out earlier, Snapshot also improves customer consideration and customer retention for key segments in the destination era.
Snapshot also brings us a lot of data. A recent press release had the graphic on the right showing that in six years we have accumulated 10 billion miles worth of driving data. It puts our traditional data, the data we use to resolve our traditional product, an entirely new perspective. Just remember, it's not necessarily true that just because we have a lot of data, we'll be able to price better. We believe we will be able to. To understand why we believe that, just think about our traditional product. How do we get so much out of such a relatively small amount of data? It starts above all else with great people. Our great analysts look over years of history for tens of millions of vehicles, for dozens of variables, and over several line coverages.
What they do is they constantly look for new relationships amongst the variables we already use to price today, and they look across the landscape, both within Progressive and outside of Progressive, for new variables they can add to the mix to continue to push us closer and closer to perfect pricing. You can see that the usual technologies we use to handle this data struggle to keep up, even as we scale them through 2013. The amount of time it took our analysts to test their hypotheses grew at almost the same pace as the Snapshot data itself. In 2014, we're introducing new tools and technologies that will cut that time by 90%. That's right. Our analysts will be able to evaluate their hypotheses ten times faster this year than they could just a year ago.
We'll learn faster, and we'll improve our product segmentation faster than we ever have before. Let's change tracks a little bit and visualize how Snapshot changes what we know about the vehicles we insure and how changes we're making to the Snapshot device will allow us to take this to an entirely new level. This dot represents a vehicle without Snapshot. It turns out we don't know that much about it. We know its make, model, and year. We know where it's garaged. We know which coverages it has and which limits and deductibles it's chosen. Once we add Snapshot, we start to get more information. You see a tail coming out of the dot now. That's data that comes from the accelerometer that we have in some of our Snapshot devices, telling us a little bit about changes in direction and speed.
All of our Snapshot devices capture speed and time of day. Here you see the speed of the vehicle captured in the color of the dot, and the time of day shown in the graph in the bottom left. This is the data we use today to achieve all that segmentation with Snapshot. Recently, we started testing devices with GPS enabled in them. Employees have had them in their vehicles for several months. One of our employees gave us permission to use the data from his vehicle to give you an idea of what doors this opens for us. Here you see one of his trips to work here in Mayfield Village, Ohio. Here you see 50 trips. Can you imagine this data for millions of across the United States? We can, and we started to ship GPS-enabled devices to our customers in April.
It's important to note that GPS was part of our original conception of Snapshot, and we took it out because six years ago, customers just weren't ready for a GPS-enabled device in their vehicles. Now 60% of U.S. consumers have GPS on their mobile device. They're ready, and we find that when we offer GPS-enabled devices to our customers, they have the same take rate as those devices that don't have GPS in them. Once we know where the vehicle is, we can bring in all kinds of external data sets to ask new questions. Here, we're looking at the speed limit of the roads that the vehicle's traveling on. Now we can start asking new questions. Does it matter what the speed limit is of the road that the vehicle's traveling on? Does it matter what the vehicle's speed is compared to that speed limit?
We can bring in external crash data that's publicly available and ask ourselves, does it matter how often accidents occur on the roads this vehicle travels on? These are just a few questions we could ask, and probably the most obvious ones. We don't think they're going to be the most insightful ones. We're excited because these are questions we couldn't even answer just a year ago. I hope you'll agree that the combination of great analysts and new tools and technologies is something that will allow us to keep our foot on the gas as we increase the distance between us and the competition.
I'd like to provide some additional color, some additional background on some key states we have in personal auto. These are states that we've discussed in previous investor relations meetings. I think you'll see each state has a unique story. Let's start with Florida. Florida's our largest state. We have $1.8 billion in written premium over the trailing 12 months. That represents an 8.2% growth in premium. The combined ratio over the trailing 12 months is 89.3%. Let me add a little bit of context. We're earning into some rate decreases. You would expect from an actuarial principle that the combined ratio would increase over time. Let me add some context to that. Year to date, our combined ratio is 92.3%, three points higher, that's working as you would expect. The market share here is 13.1%, which compares to 8.5% we enjoy countrywide.
The key issue in Florida is the 2013 personal injury protection statutes. They were intended to help lower loss costs, from a results standpoint, the loss costs have come down comfortably in the range of our expectations. The story would be so far so good. As we know with PIP, past results may not guarantee future performance, we'll be watching that. From a key action standpoint, in 2013, we lowered rates 8%. That was in response to our indications as well as to the statutory obligation to do so. One comment I'd make here is that we've grown top line by 8.2% having lower rates. You can infer there's been some unit growth in the state, that'd be exactly right. I mentioned earlier, PIP can be volatile.
I just wanted to characterize, the PIP statutes, which have helped to lower loss costs, there are constitutionality challenges associated with those. The key point I'd make is, to the extent to which the loss cost provisions in those statutes are eroded, we would take the requisite action to maintain our 96%. Again, I mentioned that PIP was volatile. Well, I think many of you will remember from two years ago, we had an issue with PIP reopens. It was actually something across the entire industry. The comment I make here is that when that opened, there was quite a bit of uncertainty. There's less uncertainty now about that from a status standpoint than where we are today. Finally, I want to share an exciting opportunity. We have just brought the bundled offer to Florida.
ASI is our underwriter in Florida, and we've been able to bring that to the agency channel in late 2013. Just a couple of weeks ago, we made that same offer out to our direct customers. We're really excited about being able to make this offer to both our agency and direct customers in Florida. Let me move on to Texas. Texas has $1.3 billion in trailing 12 written premium, which represents a 4% growth. The combined ratio here is 96.0. If you had a chance to read the release from yesterday, you noticed that we had some cat activity in the state. In fact, the CR has been quite a bit lower than that, but has just kicked up in the last month. It's 2.3 points higher than the year previous.
The market share here is 8.7%. The key issue here is growth. We'd love more of the Lone Star State. From a key action standpoint, we lowered rates 3% in 2013 in response to our indications. We've increased rates in the first quarter of 2014 in response to those indications as well. We will continue to monitor the rate level to ensure that we can grow as fast as we can at a 96. On the demand side, an interesting story here, there's a modest amount of disruption, not dramatic, but a modest amount of disruption in the home insurance market. There's been three years of rate increases, increase in deductibles, limiting of availability. As a result, there's been some more shopping in the marketplace. Our response to that has been to increase local advertising to attract more of those shoppers to Progressive.
Let me move on now to New York. New York, we've seen 14.8% in written premium growth. We're now at $758 million in written premium. Our combined ratio here is above 96. It's a 96.7. It's pretty fancy math I just did. We feel great about where we are in this state. We really like where we're positioned. Let me tell you about what that is, before I get there, just want to make one other comment. We make internal estimates about the effect that extraordinary weather would have on our results. In New York, in the 96.7, you see there's about a point of extraordinary weather, mostly from the hard winter that we've experienced. The market share here is 6.9%. That's associated with the fact that there's a strong, sizable competitor in the marketplace.
You can infer that we've seen some aggressive loss trends, that's the other key issue, from the 96.7 that we 12 months. Let me share with you the actions we bring to bear on these issues. First, we've been raising rates over the trailing 12 months, about 5%. Second of all, underwriting. Underwriting is important in all states. It's particularly important in this market. By underwriting, what I mean is less a question of accept or reject, more a question of ensuring we have the right information about each risk. I mentioned here we are validating proof of prior insurance. This really is something that would go beyond that one variable across many of our rating variables. Oftentimes, the insights we get about underwriting in New York help shape our overall countrywide underwriting agenda.
Lastly, I want to make a point that hearkens back to one that Glenn made. We do product model upgrades on a frequent basis as opposed to large, discrete rollouts of wholly new products. In New York, in second quarter of 2013, we improved our segmentation through a product model upgrade and also reduced complexity. It's the combined set of actions here which give us a lot of optimism about what we're doing in New York. Last slide for me, Michigan. Here we've seen 12% growth in written premium. It's 14% on a non-ceded basis. Trailing 12 written premium is $572 million. Our combined ratio here is 94.9. They also suffered through a hard winter. Our internal estimates of weather, we'd say it's about two points on their combined ratio. The market share here is comparable to what we see in countrywide, so we're at 8.2%.
The key issue in this state is one that afflicts not just us, but all the industry participants, and that's the unlimited personal injury protection. Let me share some context. Out of every $100 of auto insurance premium that an average Michigan insured will spend, 40%-50% of that will be for PIP. It dominates the economics. From a key action standpoint, we've raised rates over three years, over the last three years, to the tune of 28%. That's in response to trends and indications. Our competitors have responded more recently, and their more recent actions have improved our competitive position, giving rise to the growth that you see. Also, we do several product upgrades over the last three years. This also allows me to comment on another thing.
We've had the state product management structure, which allows the product managers to make observations and bring to bear not just countrywide insights about product, but also local nuances. That's true in Michigan as true in all of our states. Lastly, there are some contemplated legislative changes being considered. We are participating through the trade groups on those discussions. The key point here would be whatever the rule set is that's finally designed and defined, we will focus on profitable growth within that rule set. Last thing I'm going to mention, this is actually not on the slide. I just want to share some exciting news with you. We are in the Massachusetts agency market. We entered a week ago, on May 9th. We entered the direct channel six years ago, and we are now in the agency channel.
I'll tell you why we're excited for a lot of reasons. First of all, the agency market in Massachusetts is $3 billion, it's a sizable market. It's also a bit of a milestone for us. This is, of all the 51 jurisdictions, we are now in both the independent agency channel as well as the direct channel. This milestone has been greeted with quite a bit of celebration within our walls.
All right. Quick update on our commercial lines business, which continues to be an important contributor to overall company performance and profitability. By focusing on commercial auto as a primary business, we have been able to outperform the industry over an extended period of time, both in terms of bottom-line profitability, with a combined ratio ranging between 8-15 points better than the industry, and in most years with top-line growth that exceeds the industry average. 2013 was not one of those years for top-line growth, and contributing to that were some corrective actions we needed to take in our for-hire transportation segment, which I'm going to talk a little bit more about in a minute. What's really at play here is that the industry continues to operate in very unprofitable territory, with only slight improvement in 2013.
We believe much of the industry is now dealing with a substantial headwind of adverse development in the most recent few accident years. We had some adverse development on the most recent accident year, I'm confident we've addressed that in our reserving and in our prospective pricing. However, as long as the industry continues to operate in this territory, we may, if necessary, have to forego some top-line growth to ensure we make our targeted profit margin. John and Glenn and others have shared with you a segmented market view of personal auto. We take a similarly segmented view of commercial auto, and the diagram here depicts the 5 high-level business market tiers, or BMTs as we call them, that we focus on. This construct, which is admittedly a simplification, is useful in a number of ways.
Perhaps most obvious for marketing, as these different BMTs, they're defined by the nature of the businesses that we're insuring. These different BMTs consume different media. They have a different propensity to buy direct. They have different needs. In the independent agent channel, they tend to access different distributors. Yes, it does help with marketing. Much more important than that, we see very real differences between these BMTs in things like loss cost trends, loss development patterns, retention characteristics, and even the transactional intensity of the policies and the cost for us to service the customer. All of those things are important inputs into our pricing. By taking this focus by BMT, we're able to price more accurately and have greater dexterity in terms of how we respond to changes.
Beyond that, we know that certain macroeconomic factors affect commercial auto insurance, what we see is they affect these BMTs differently. Whether it's just in terms of vehicle utilization, which ultimately drives frequency, or just demand for the product on the aggregate level or at the policy level in terms of the coverages purchased and the limits carried. As I said, I did want to talk about one of these BMTs in a little more detail, and that is for-hire transportation. This is an important segment for us and one we have invested in over the last several years in terms of adding coverages and limits and building specialized claims capability to service the customers. We've also grown this segment pretty substantially between the years 2010 and 2012. That growth was by design and consistent with the investments we made.
That growth also occurred during a pretty interesting time for the industry. What was happening is the economy was beginning to come out of recession. As the economy recovered, the amount of freight going over our roads and highways started to increase. That is captured on the chart on the left, which is the American Trucking Associations' truck tonnage index, and you can see that going up. As that went up, our bodily injury frequency also started to rise, particularly early on in the recovery, where it accelerated at a pretty rapid rate. That showed up in our loss ratio. We've addressed that with our pricing. We have raised rates in this segment 73% over the last 36 months, and we are very comfortable about our current rate levels.
While frequency was going up, we also saw increased enforcement from the Federal Motor Carrier Safety Administration. That's a good thing. More roadside vehicle inspections were occurring, and more operators were being cited for things called out-of-service violations. These tend to be fairly serious violations that require the vehicle or the driver be taken off the road until the violation is remedied. There was a secondary effect of this increased enforcement, and that created a little bit more shopping in the marketplace as these drivers and operators were experiencing rate increases, or in many cases, having their policies canceled. They were back out there shopping. As much as it kills us to ever be out-segmented on any variable, for a while, we saw our new business mix shift towards more operators experiencing these out-of-service violations. That, too, showed up in our loss ratio.
We've addressed that through our new business underwriting. We've also addressed it in our renewal book with targeted price increases and, in some cases, selective nonrenewals. We feel good about this segment right now. On a trailing six basis, the combined ratio is at our target and trending favorably, and we're in good shape. Last point I'd make on for-hire transportation is that this is still very much a hard market. Rates are continuing to go up. Competitors are placing restrictions on business or withdrawing altogether from the marketplace. Applications for transportation risks to the residual markets are increasing. In insight we have the national residual market servicing carrier. This is a hard market. We will have an opportunity to grow this segment profitably. I'm sure of that. I just can't be sure about the timing.
I'm going to return to this segmented market view for just a minute because not only do we see meaningful and actionable differences between the BMTs, they each represent a different portion of the market. In each, we tend to have a different market share position, and therefore, different opportunities going forward. In the three smallest BMTs, we actually index higher on market share, double-digit market share. This actually makes some sense. These BMTs tend to be vehicle-centric BMTs. By that I mean the vehicle is central to how the business generates its revenue. Their insurance spend is disproportionately towards auto. More than 80%, in most cases, of their insurance spend or their insurance budget is allocated towards auto. It kind of makes sense we do well there.
The two other BMTs, contractors and business auto, collectively make up about 75% of the total market, and we index slightly lower on market share here. These are not vehicle centric businesses. Other commercial coverages, commercial liability, commercial property, workers' compensation, begin to have much more importance. Our own research tells us that it's these other coverages. For contractors, typically it's general liability. For small business owners, it's a business owner policy. These other coverages are the ones that drive the placement decision, whether it's a direct prospect or whether there's an agent involved guiding the decision. The chart shows some representative small businesses and how they allocate their insurance budget to these different coverages. These small businesses are all businesses we really like from an auto perspective. We'd love to have more of them.
As you can see, commercial auto represents, in all cases, less than 50% of their total spend. Our influence is reduced. What do we do about it? One of the things we've done a few years ago was we created Progressive Commercial Advantage. Progressive Commercial Advantage is an in-house call center based agency where we work with other companies to create small business commercial insurance packages for our direct customers and direct prospects that are coming to us seeking these coverages. The agency is still pretty small, but it's growing rapidly, particularly as we fill out the product map with more offers. We're encouraged by two things. One, when we create a commercial package with our auto, we are seeing early but very significant gains in auto policy retention, gains that will lead to extended PLEs.
Two, the brand strength of Progressive is already generating a significant number of prospects coming to us looking for these other coverages and therefore creating auto opportunity for us as well. That's working well for our direct customers. Of course, the bigger opportunity here is in our broad-based independent agent distribution, because this is still an agent dominated business. To capture that opportunity, we know we're going to need a slightly different solution. What we'll need to do, basically, is take the capability of Progressive Commercial Advantage and deliver it to the desktops of our independent agents in a smart and efficient way. We've asked our agents what they think of this concept, their response has been very positive.
What they've shared with us for the small business customers we're interested in, it's these other coverages that are oftentimes the hardest and most time-consuming for them to handle, frequently having to access wholesalers and other distribution outlets to get it done. We believe we are somewhat uniquely positioned to create a robust small business commercial market and deliver that market to the desktops of our independent agents to grow our business. It's a major priority for us and something you'll hear more about as we achieve particular milestones. That concludes my commercial update. It concludes our lightning round. With that, I'm going to turn it back over to Glenn.
Pick up where I left off with the sort of a business model. We're acutely aware of what we're talking about today. We want a segment of customers that everybody else wants. We're acutely aware of that. The real question is, what's our business thesis? What gives us a right to think that now is the right time for us and that we have the right assets and can build on the skill sets that we have? Let me hit just a few of them that have come out today. We are a leading producer of pre-Robinson customers. You can't retain what you don't have. That's critically important. We haven't talked a great deal about our other strategies around Diane's and so on and so forth today. We're focused on one very significant but future positioning of Progressive.
We have those customers, not as they exist today, but as they will look like years from now and decades from now. Brand relevance and recognition for future Robinsons. We are a contemporary solution for these people. We know that. We do a lot of research to determine. We have what they want. We are no longer comfortable being just the prep school. We're going to go deeper into the insurance relationship with our customers. If that means providing products that we haven't previously provided, so be it. We will still be very much an auto centric underwriting company. The other products you've heard about today, it's that bundling notion.
My last comment with regard to assets, this has been proven now in numerous ways for us, it's also very apparent in the rest of the industry, and that is consumers are very accepting of an aggregator who can meet their needs and provide coherent customer experiences. If those products happen to be OEM labeled differently, so be it. When I'm talking to groups inside Progressive, I often say, how many of you have a printer for your home computer that's a different brand? You can just internalize that, the numbers are very large. Because the solution works and because people associate best of breed characteristics, they're comfortable with that solution. There's lots of other models of distributors, Amazon, so on and so forth, that do this sort of thing.
We want to make sure that we start a relationship, and we're not willing to give it up. Now that we've earned the right with our brand and opportunities that we've talked about today, we want to go deeper. Skills. We've got to make these brand consistent experiences. This can't be just a collection of stuff. When we have things that will be primarily product related, such as we have with our property partners, where we've done a single deductible. No customer expects when their car is vandalized that that's two policies that come into play and two deductibles that come into play. We want to make that one deductible. We want to make the product more intuitive. We want to upgrade and update the product and the product experience. We can do that without necessarily being the manufacturer. Brand consistent provider relationships.
We've learned a lot about having the brand and someone else's checkbook. It's not always the most convenient marriage. We've got to be very clear about those marriages that will work and that will won't. We've learned a great deal. The final three, you've got all these in your book. We've got to use the same kind of segmentation skills that we've talked about in several different areas today to segment our customers, then build on that marketing skill and start to become as good a marketer to our current customers as we have been to the broad market. That's going to be a real skill. There is so much marketing out there today that frankly turns me off because they're talking to me all the time. They're spamming me. It's not elegant. It's not relevant.
We've got to make our marketing to our current customers really elegant, really relevant. Our customers, we know they don't want to form the greatest relationship with us. It has to be the relationship at the right time, right place, right product. We're going to need a lot of life events products. What I know about wedding insurance would fit on the back of a postage stamp, and I plan basically to keep it that way. If that's what Diane needs at some point, and if she places personal utility on that, and our currency is policy life extension, I'm all in. I don't want to write that. If we can find it, bundle it, and present it in a great customer experience, that's the new model of Progressive. We have those assets today. We have a lot of things that are supportive of it. Customer data repository.
You've seen a couple of screenshots and different examples of customer summaries and so on and so forth. We're good. We're probably not good enough. We already have plans to actually extend our customer relationship management systems, and we'll be doing that as the year progresses and into next year. That's a capital investment that we're willing to make, and it seems like the right time to make it now to support this strategy. John talked a little bit about the in-house agency. Clearly, that gives us the opportunity to use multiple carriers and be able to get greater conversion. It also gives us the opportunity to test some products. Some of the new things we'd bring in, we're not going to sort of make the investment necessarily to have them be total plug and play right away, but we can test some things in an in-house agency.
Last point is, we will need to make our online experience, since we've said that that's really a place that we play in, and that's what customers expect from us. We'll have to have a lot more plug and play capability where when you come to our website, it doesn't feel like it's the collage of six different websites, but one website, one experience, great plug and play capabilities. I'll say it one more time. Our benefit is all about policy life extension. We're very good at what we do, we think. I have no interest in changing the focus of Progressive away from manufacturing auto insurance. The property coverages that we've touched on today are the lion's share, the lion's share of what we need. Other than the renter's comment that John made, we will likely find people that are willing to work with us.
Over the last couple of years, we've answered a question for ourselves, are people willing to work with us? The answer seems to be yes, because we have a brand and we have the opportunity to present it to them. This is a very exciting notion for us to sort of think about taking our customers on a much longer insurance journey and get paid for it simply by policy life extension. Brian and I will take your questions now as we go into the final part of the session.
Hi. Cliff Gallant from Nomura. Up here. When the customer goes and gets the bundled quote, what % of the time do you expect to win? Will it be the same as when they get the quotes from others on the auto policy? If they are winning, how do the economics work? I understand how a State Farm, for example, can offer a discount when you're getting multiple products, but the economics when you're splitting that with a different underwriter, I would think would have to be different.
Two responses there. First is sort of the bundling that might occur at new business when they're actually buying more than one product at new business. Don't assume that to be a particularly large % of what we're talking about today. It may be that they start their journey, which we are in the advantage position. They typically start with auto. It's more a matter of adding those things on. With regard to sharing indemnity discounts, frankly, as I said, with a brand and working with customers, there are opportunities for us to have those negotiations, and we are able to take a multi-product discount. We've actually offered a multi-product discount before on something like homeowners, even within our auto, independent of whether we had any relationship with the homeowner carrier.
We've also seen situations where the homeowner carrier we're working with now will actually extend their own discount, and that, for us, is exactly what we want to see happen. Frankly, we want to get that even closer together, which is why we're forming tighter and tighter relationships with some of the homeowners companies we're dealing with. One of the reasons that, frankly, in the renter's case, we decided that that's a product that we can very clearly do ourselves. We may have trouble seeing you just from the light, so make a little movement.
Hi, Ian Gutterman, Balyasny. If I can ask two, I guess start with the easier one on homeowners. As you get bigger and bigger in this, how do you manage the service experience? Tricia talked a lot about that on the auto side. I think of all the things you've done over the years on service centers and so forth, it seems that homeowner, if you're going to have a third party do it, and especially if they tend to be smaller than the national players you're competing against, you probably can't ever get it to the level you want it to be. Where it's going to be as good as the auto. Is that fair, and how do you manage that?
I think it's very fair to start with that sort of assumption. Things that concern us a great deal about brand delivery, brand experience in claims, all of those things are very real. Whatever we've said today, don't assume that we think it's falling off a log.
Sure.
Having said that, we have a level of negotiating power, if you like, I don't want to say that because these people are very important to us, to make sure that we actually have some agreed-upon specs.
Okay.
In some cases, don't take this out of context. In some cases, we may start to do some of the servicing ourselves. Some. That probably isn't necessarily the solution, but it might work in some cases. Just an example to that, we're certainly working with some of our homeowners companies to suggest that in the case of a catastrophic loss, that we could at least be available for some of the claims reporting. We may not have all the skills. If I was taking my agenda of skills there are probably some skills that'll be added over the next few years. In short, your concern is our concern. It's not going to be perfect. We're only going to do business with the people that we feel that we can both audit and review and will hold themselves to the same kind of standards.
Some of those we'll miss, no doubt. We will be, I think, in a very different place five years from now and have a very clear view of what it takes to make this happen.
Great. The other one was you talked about the high-limit auto customer and adding the Umbrella and how important that is, then growing them into a homeowners customer. That makes a lot of sense to me. I guess my question is how viable is that for Progressive? Just when I think of a simple message. It's Name Your Price, which tends to get people to buy lower limits. It seems that Allstate's taken Esurance to try to capture what you're talking about is get someone who's comfortable being technologically savvy, but buying more coverage. Is it a fight to get in the right marketing position? Do you need to change your ads? Do you need to change other messaging so that people don't think that Progressive is the price gun and buy low limits?
Yeah. I think I followed most of that, and if I don't respond directly, please redirect. I do think you made one assumption that isn't accurate. Name Your Price. You should not assume that people gravitate to the lowest possible price. In fact, what we've found when we were doing some control testing that obviously you don't controls forever, is that people actually would start with the recommendation, move, but the vast majority ended up buying based on the same recommendation that we made relative to their requirements. It is not a fair assessment to suggest that the Name Your Price concept has gravitated people to a lower limit pricing. In fact, we study lower limit pricing in the marketplace, and without naming names, we're not the low-limit pricer.
Glenn, a couple of years ago, you were very excited about the Snapshot usage-based insurance in terms of the pricing ability relative to your prior models. How has Snapshot developed relative to your expectations over the last couple of years, the profitability of the customers, the policy life, and anything else, the growth or the uptake?
I think I can hit those pretty quickly. I would say that relative to our a priori, no database sort of expectations, the average discount is slightly lower than what we might have expected. Average discount. Obviously, the range is still what we're there. The retention benefits are absolutely there. We don't want to start talking about Snapshot as if it's a separate book of business. It's part of a big book of business. When we isolate, as we obviously do, for tenure and retention differences, Snapshot is worth a significant amount. Relative to profitability, we compare it to a cohort that is largely similar in terms of inception, but not taking the discount. Our goal, and we're achieving the goal, is to have a combined ratio of parity.
If in fact their average discount is X, then it has to actually be totally deserved, and therefore, you'd expect the combined ratio to be the same as a book of business that didn't have the discount. Actually, we're really pleased with it. I make that first comment that it would have been great to think that the average discount was 25%. It didn't work out to be that much, but it's unquestionably significant and important. We're not ready to even talk about the details, but the kinds of things you saw from Steve, we're always working on new models and new incorporation of variables. Snapshot shouldn't be thought of as it is what it is today.
You should think of that in the same mode as our GE curves that we'll always be trying to push it out and look for more models and different implementations of Snapshot in the future. Couldn't be more happy on those profitability tenure and an uptake. I report on that fairly regularly on our conference calls. The uptake has actually continued to move up, and now we think it's at least at or maybe even above that segment of the marketplace when we first surveyed and said, "Yeah, I'd be willing to try this." We're probably actually into the part of the marketplace that said maybe. We'll see a little bit more. My projection is we're going to see a little bit more activity from the rest of the marketplace. Remember, a few years ago, we announced that we would actually license Snapshot to other companies.
That actually gives them the availability to be in market next year at about this time. We have one more year before at least people under our license can come into the marketplace. As that becomes more of a mainstream part of the consumer psyche with regard to auto insurance, we expect to benefit disproportionately because we write more new business disproportionately. This is fundamentally a new business issue.
We actually have some questions from the webcast folks, one of them relates to Snapshot. I'll read that, and we'll answer it. How's that?
Good. Great.
Glenn's letter indicates a 31% growth in Snapshot. Overall, the company is seeing 7% growth in direct and 1% growth in agency. I believe that's PIFs. What does that mean for growth rate for non-Snapshot business? Is there a cannibalization effect? How does one square these numbers? Couple questions in there.
Sure.
The 31% growth, keep in mind, it's the aggregation of both new business coming in, and our take rate on new business, particularly in the direct channel, is very healthy, very high. Something like a third of customers are taking Snapshot as part of the acquisition process. Also the growth is the renewals of customers that started Snapshot a year ago or two years ago, and that continues to renew and actually has the higher PLEs that John referenced in his discussion. It's a combination of continuing new business acquisition via Snapshot plus the retention of previous Snapshot customers that's following that growth. Certainly more so on the direct channel than the agency channel. I would not characterize it as cannibalizing our other business. I think it's a rating variable. We believe in a segmentation value. It's going to be even more accurate pricing.
At those rate levels, we feel comfortable with acquiring those customers and retaining those customers, and we think we're going to keep them for a long time. I would not characterize Snapshot as a cannibalization of our business. It's another very significant rating variable.
Hi, thank you. Josh Shanker from Sanford Bernstein. Most interesting statistic I saw was the 40% of your young Robinsons are signing up for Snapshot. That seems helpful. The question I think would be, what can you guys do or what are you thinking about doing to actually make this more of a new business driver among that population as opposed to sort of just something they sort of passively respond to in the context of once they've already decided to become likely your customer?
Just the last part of the question again, Josh.
Oh, the question would be is how can you get them to pick up the phone and call you?
Oh. Yeah. We hadn't thought of that.
Yeah.
No, in all seriousness, we continue, and I think I've said this several times, so I wish I had different words or we'll continue plugging away at it. When you go first with something of a new technology like this, there's a lot of barriers to break down. I really can't sort of grade whether we've done extraordinarily well or not because there's no way to grade that. We're going to keep pushing. We've tried, as you've possibly seen, different angles on our television advertising. We've even gone outside of the Superstore and gone with what we called a rate sucker campaign, where we could sort of appeal to people's thought of, well, you might be paying for someone else's insurance, which seems to be actually from a testing perspective, a positioning that really worked well.
We're going to keep doing those positioning type statements and keep finding ways to make the message of Snapshot resonant with consumers. I think at 40%, especially of this target market, we're starting to get there. We're starting to get there.
I just appreciate you guys teasing us with the chart of the GPS and all the cool graphics. Is there any way that you're willing to share with us how you think the consumer sort of offering is going to evolve? I mean, do you have more data in there? Is there going to be different discounts? Is it going to be continued monitoring? Anything along those lines?
Stay tuned.
Okay. Thank you.
Good try.
I will say what Steve said. Recognize we did from the early 1990s, we were in, so yes, no one was talking about it then and we were probably considered crazy or whatever, but we were doing a lot of this. We had GPS in the earliest research that we ever did. GPS is not new to us. We took it out very deliberately to try to find a packaging of something that would relate to the consumer in the time period that we thought made sense. It's very hard to know whether we were right or wrong on that, but GPS was the one we took out. We've always known the power of it, and I'd also suggest to you don't have to compete against yourself. If you've got something that frankly is additive, then you can wait and add something more later.
Some of what we have in store are things that we've known about for a period of time, and we're very confident that it makes sense for us to be additive. Staying one step ahead is largely a lot of what we try to do.
It's Paul Newsome with Sandler O'Neill & Partners. Could you talk about the philosophy you have towards your internal agency division from a revenue perspective? Do you consider it a separate profit business or is it an accommodation? How do you think about, frankly, the money you spent to build those businesses returning to you?
Sure.
Yeah.
I think how we think about it primarily is if they're able to sell the other products and the PLE extension on the auto, that's the primary win. Certainly, by selling the other products, we're also going to get a commission stream for other products alike. That helps fund the investment that we have to make in it. We're more than willing to make the investment in systems and certainly people to get it up and running, knowing that we're going to have a commission stream to offset some of those costs. The real gain to the company is in that PLE extension on the auto, and we think it's going to more than pay for itself. I'm very confident of that.
Theoretically, you're thinking of both the gain plus whatever revenue you get.
For sure
in total as sort of your return on that internal agency.
Sure. For sure.
Thanks. Vinay Misquith from Evercore. The first question is on pricing. It seemed last year you were overrunning a little bit, and you seemed more willing to raise pricing slightly less than loss cost trends to really push up the top line growth. You have succeeded. Texas and Florida rate reductions, and that's helped with PIP. Where do you think you are right now with your profitability, and with the wind at your back sort of receding now that the rate decreases are gone, should we see a slightly slower pace of PIP growth in the near term?
You'll have to do the translation to PIP growth, let me sort of answer pricing philosophy. You're right, in a couple of large states, I signaled that even after taking rates in 2012, there's always an adjustment. I want to try to sort of take just a second to realize there is never a time that we think we have the rate exactly right. There's data coming in every day. The question is, what are the intervals between adjustments? Which can be disruptive to consumers, disruptive to agents, but we're always trying to price to trend. The reductions that you saw weren't sort of, gee, we're sitting around saying we need more volume. They were reductions because they made sense relative to the new data. Specifically in Florida, you heard Sanjay talk about the PIP effect.
Very hard, even though a lot of rhetoric will go on before a law is passed about how much reductions are expected. Unfortunately, it's the people who are accountable that are the only ones that have to add them up at the end. We're a little cautious to adjust. We'll adjust where we think is necessary, but if we see some trends relatively early that are either better or worse, we'll adjust to them. In Texas, you actually saw a few noted on Texas, we took rates down, and then we took them back up again. That's going to happen too because things change. I think the bigger question is where do we see ourselves now? We went through a winter where there was a great significant increase in collision frequency. That's not news to anyone here.
Surprisingly, when you have a high collision frequency, somehow that seems to offset bodily injury in a strange way, that those cars are not on the road and doing other things. We've seen a little bit of a change in the last four or five months. I'm not sure we're at steady state if you take our last four or five months and say, let me try to find the steady state. I think we're priced as John tries to get our personal lines business, and John Barbagallo our commercial lines business. We try to price to 96 or better all the time. 96 is really sort of the point estimate we're trying to price to. If we're going to be off, we're going to be off on the downside of that. It's not a question of sort of out there seeking growth.
We're not going to sacrifice profit for growth. However, I suspect, this is clearly and a somewhat informed decision, but not necessarily, there's no perfect decision. I suspect we're going to see a longer period than I expected of relative rate stability for the consumer. I suspect it's going to go a little deeper into this year. If I was giving sort of any guidance relative to Progressive, that our rate level will probably go up somewhere between 3% and 4% for a year. Which is certainly consistent with what you might expect in other consumer products and so on and so forth. That's all driven by medical costs. The Affordable Care Act is throwing a little bit of an issue in. There's all sorts of speculation of what that might mean, positive and negative. We're going to just wait to see the data.
there are variables that could play into that, but if I had to sort of pin one down today, I would say expect relatively modest rate increase for the rest of the year.
just to further clarify, through April, we have raised rates in auto about 2%. We feel we reacted to data and raised rates modestly. We think taking those smaller bites of the apple, like Glenn has often referred to, is much better than the end of 2012, when we were raising rates 5%-6%, in some cases double digits. smaller bites of the apple so far this year, 2%. frequency has been very different throughout this year than some of the past years. Frequency was up a lot in the first quarter, PD and collision in particular. April, frequency was down.
Sure. My second question is just on capital management. You raised about $350 million in debt recently. Your debt to capital has been low. Think of this as more used to buy back stock, or what are you planning to do with this debt? What is your normalized debt to capital? I thought it was close to 30%.
I would not say that as the normalized debt to total capital. Think of the 30% debt to total capital is sort of our cap, there have been times when we have gone above it, we think that more as the cap. Right now we are about 25%, frankly, the debt issuance was capital position strong. Take opportunistic, take advantage of low interest rates environment, lock in some long-term debt for 30 years at 4.4%, I am pretty confident over the years we are going to find good use for it, nothing more than that.
Thanks. Meyer Shields, KBW. Can you talk a little bit about the resilience of the expected lifetime earned premium growth, relative to actions that competitors might take, whether it's cutting prices, increasing advertising, doing some of the things that you're doing?
Sure. Let me take a shot at that, Brian might be able to add on. When I introduced the notion gave you a quick example, I also said simple examples can lead to some very awkward outcomes because we intend, by our own actions, to extend the life of a customer. We will almost certainly take some rate adjustments during that period of time, as you mentioned, the marketplace will change in competitors. The answer in some sense is impossible. For retained business, if we keep, as Brian just said, smaller bites of the apple, our ability to retain our customers, because you have to give them reasons to just stay and be somewhat inert around their rating. If you give them, let's just make a number, 10% rate shock, they're going to go out and shop.
Even if a competitor were to take a rate down lower, the real issue is have we given our customer a reason to go out and look for that? Which frankly may not be a good thing. If it's 10% lower, it may not survive very long. For the renewal book, I think the resilience is very high, when you see some of the references we make to other competitors and their tenure length, when John said Robinsons, hey, we'll put four because we can stand behind that number with the plus sign. We know what it is in other companies. It's 60 months. It's 40 years.
The resilience is very high consumers are not as much as we're excited about auto insurance, most consumers are not so excited that they go home at night and say, "Gee, I think I might just go out and shop my auto insurance." Failure to give them a reason to do that is actually a big part of our strategy. I'd say very high on the resilience, shifting to the new business, it's always a dog fight, we're pretty good at that.
The other thing I'd add to that, I think there is a little bit of difference in terms of let's call it price elasticity at renewal time based upon the market segments. The Dianes and Sams are a little bit more price sensitive than the Robinsons and the like. We do see even based upon the same rate change differences and renewal rates on that. I'd say based upon the market segments, there's a little bit difference in terms of resiliency there. But for us to continue to grow policy life expectancy, lifetime earned premium, keeping rates relatively stable but always accurate is important point of it. One of the things in terms of Glenn's this invisible balance sheet, this unearned lifetime earned premium, there can be a few components to that.
You can have more policies in force, we need to do that by keeping the acquisition engine going and writing as much new business as we can. I continue to say, I think I write more new business than any company between both channels combined. It's extending the policy life expectancy, which we've talked a lot about things we can do to do that. It is also average premium per each of those policies, whether it be rate changes or additional coverages or things like that. I would characterize it as three component pieces. Number of policies in force, the length they're going to stay with you, and then the average premium per exposure.
The Robinsons obviously have a much longer life expectancy and they buy some coverages that add to it, much more full coverage there, more multi-car there than sort of the Sams and Dianes of the world.
We're picking up on that because it's obviously the theme of this meeting, going to Tricia's example, it really is sort of that conditional PLE. They start with one, given certain actions, that PLE can change, that's really the key to this business strategy. We talk about it in terms of customer names and sort of nice actions. Really when you get down to it, is building conditional PLE conditions.
Maybe we'll take another question from the webcast.
Sure. Okay.
I'll read it, but I'm just letting you know this is one you're going to answer.
All right.
Okay. All right. Just letting you know. Is Walmart comparison rating product a threat or an opportunity? Typically, you fare better when there is more shopping, but giving customers more pricing information could bring down overall pricing. This is in the news recently.
Yeah. I actually have some personal experience with Walmart's interest in auto insurance over several years. I won't go into that history. What they've done is gone with a comparative rater in store. We're represented on that comparative rater, and so it's really to be seen whether people are willing to sort of take an in-store promotion and then go home and actually do something. I don't know exactly how this will work out, but I'm happy to be positioned as one of the lead companies. If we associate certain characteristics with the business that don't meet our criteria, we'll react to it. The answer is, better to be in than not. Walmart clearly has some reach into certain parts of society, so we're in. Other than that, we'll just see how it plays out.
The comment, typically you fare better when there's more shopping, that's fair. We're interested in shopping. Not all shopping is created equal, and we'll have to make sure that we're comfortable being represented on that platform.
Bob Glasspiegel from Janney. I'm going to throw a couple hypotheses out, and you can tell me whether I'm correct or off base on them. It seems like the direct business is a great business with better growth dynamics, better profitability, and it's demonstrating Progressive's traditional edge versus the industry. On the agency business, where Progressive's roots have come from, I feel like the edge versus the industry, both with respect to growth and profitability, has narrowed over the past 10, 15 years. I'm wondering whether you agree with that, whether you think rate aggregators are an issue on why the edge has narrowed, and why can't this business demonstrate traditional growth characteristics and margins versus the industry traditional with where Progressive has been?
Is it just this is not as good a business as it used to be, and you just have to battle harder and fight for narrower margins? Has Progressive slipped a little bit on execution versus its competitors?
Let me take a shot at that. Obviously, I'd be reluctant to say we have slipped in execution because I don't think that's the case. You're right, the direct business is a great business, the agency business, and we're a very large player in that. I think the dynamic that gets at your hypothesis is really the comparative rating mechanism. To the extent that Progressive, and I still believe this is very true today, has a strong relationship with agents based on technology, based on great service, I think. Test me out. I mean, go to a group of agents and ask how they feel about Progressive servicing their customers' claims and policy service and servicing them. I think you'll get very high marks, and we've shown that in a couple of situations in these meetings.
What has changed dramatically in, let's say, over the decade, is the prevalence of comparative raters. It's not always the agent principal doing the rating. It's other people. Oftentimes price has become the single most important factor in the agency. We're going to price to our profit margin, which we have consistently achieved. I actually give us high marks on we're going to stay and do that. The growth is a little bit more distributed, and I tried to make a comment in my annual report letter to shareholders this year to recognize that at any given point in time, someone might be taking rates down, up. It's always a changing situation, people only shop on one day. The comparative raters have probably created that auction environment that we've referred to before. I don't think we've lost anything.
I think we've attempted to position ourselves on comparative raters in a way that's attractive and advantaged for us. We love the channel, and I'm more than conscious of the growth rate in that channel. I'd love to see it higher, and I think there'll always be a level of competition there where people are dependent on that channel and going to try to get their growth from that channel. I expect tough sledding there, I also expect we'll get a good amount out of it, even if it turns out to be the case that we grow faster in direct.
I'm optimistic about the agency channel. There's no reason we cannot win in that channel. We've got good cost structure. We've got good pricing segmentation. I think an opportunity that you've heard a lot today, when a Robinson comes into the agency office, I don't think agents have looked to us first. If they can get in the mindset, Progressive Auto, ASI Home, bundled package, there's no reason they can't write those customers with us. We will grow our auto preferred business and agency channel with a bundled product in that agency. I'm very excited about the opportunities in agency in terms of expanding the distribution of the ASI product, adding it to more agents. I think there's lots of upside, and our upside is in that Robinson market segment, and there's no reason we can't win in that space.
It seems like the auto CPI has ticked up the last two months. Today was a pretty big acceleration from last month. You've got to be careful looking at that data. Am I wrong to think that the industry conditions may be getting a little bit better? You're talking about raising rates. Is there an opportunity with your profitability where that is maybe for you'd have to take less rate than competitors, or you do not see that happening?
No. It'll be whatever we see in the data. I hope that's the case. My goal is not to take rates up. My goal is to always maintain margins. If we see trends and conditions that soften, and I wouldn't rule that out, I think we'll be as advantaged as anyone else. One other comment I'd make on your first one, and this is a little bit to human nature, and hopefully, I say this correctly and never unintentionally position someone I don't intend to position. With agents, many have seen us in that first era, and that's really how they associate with us and have often found the needs that they want to meet with other companies. I understand that.
When we have introduced our product into states of more recent times, I'll use the most recent one, and I think a data set of nine customers. My own caveat. Massachusetts, the first nine customers, they looked very representative of a marketplace not skewed necessarily non-standard. A lot of it has to do with the positioning that we have in the agency channel.
When John talked about bringing a renters product to agents, our research suggests that that's going to be very well received, and it's going to be one more step that starts getting the agent to say, "Maybe I can really start to build a book of business with Progressive and have that transition into PHA." Our rollout of PHA, with agents, which we've now said we'll not only expand the states but expand the distribution, should collectively start to reposition us with agents, even those that have perhaps positioned us a little differently in the past. Because when we actually do it without that history, we get a much more preferred book.
Thank you. Kelvin Pang with Morgan Stanley. Two questions. First question is that you started writing renters policy on your own book. Why wouldn't you write homeowner policy on your own book, too?
Homes don't move very easily when big winds and hurricanes and things like that come. To be perfectly frank, renters is something that, John said it, relatively low severity, and frequency will be, we think pretty low as well, but it's pretty low severity. We're doing this not because we necessarily want to write renters insurance. We want to have that relationship become deeper and stickier. Take yourself back at some age, we want to be able to build that relationship with you with two products. Tricia showed you a very simple graph that almost any financial services company can show you. The more products you have, the longer people stay. We want to get that second product in very quickly. Renters is just something we feel very comfortable about. Your extension is, well, if you're comfortable about that, why not home?
You also heard John say, when you quote any one homeowner, there's a very high non-quote return. We are advantaged, I believe, by having multiple homeowners so that we can use multiple homeowners to actually find one that fits for our consumer. That is obviously, when we're on the auto side, we don't like being compared to others and having that auction environment, but from a consumer perspective, it may actually be to our advantage to have multiple homeowners rather than one. I would tell you very clearly, we announced today going into renters, do not necessarily assume that there is a corollary to that there's homeowners yet to come.
Okay. Second question on the PIF growth. Last year had been improving quite a bit. How much of that improvements you think is coming from your lowering price or not increasing price as fast as the previous year? Now seems the PIF growth in the past few months have been kind of stabilized or stalling. What's the next leg for that growth? Are you happy with it right now?
You want to talk on that?
Could you start? Was it about the direct channel PIF growth you're referring to?
Yeah, both channel actually.
Certainly, the policies in force growth in the direct channel is now up to 7%, and last year it was much lower, and we've consistently seen it continue to grow. Some of that is due to the acquisition of new business. In the first quarter, our new apps in the direct channel were up over 10%, I think 13% for the quarter. That was part of the growth in terms of policies in force, as well as we are now seeing policy life extension in the direct channel. The combination of those is following that 7% policies in force growth. We still want to continue to get more policies in here in. We increased advertising spend in the first quarter in double digits. It was a lot in February, a little bit less in March, and certainly still increasing advertising spend in April.
We still think that direct channel has policies in force growth momentum. Getting to eight or nine, 10% will likely require either a lot more new business in or a continued improvement in retention rates. Both of those are possible, but the 7% we think is still pretty strong. On the agency channel, I think it's a little bit different story. 1% policies in force growth, we'd like it to be more. We certainly would like it to be more. Last year, we had started out in a little bit of a deficit, so we got to positive territory. We'd like to see it increase. There, I think we need to get a little bit more competitive and get more new business in the door.
We thought February was looking pretty strong for a little bit of time, but it sort of moderated towards March and into April a little bit. Some of it is the new business production in the agency channel.
Thank you. If I might add one more question. It's just April results you released out yesterday. The combined ratio actually pretty good compared what you have been done in the past few years. Is that just one month or something behind it?
I'd say it's one month until you see next month.
Right. Yeah.
Let me take one that came up on the screen here so we don't feel like we're leaving our web partners out. Is part of the problem with reaching the Robinsons that Progressive doesn't have relationships with large, high-class, independent agencies that the Robinsons would typically frequent? I think that covers most of it. It's from Bryan Perry at Sansome Partners. I think the answer to that is yes. It doesn't mean that we're not welcomed in high-end or high-class agencies, but higher end, let's just sort of say serving higher end customers, I don't want to choose between agencies, are typically agencies that have already had relationships with companies that bought them multiple products and bundle them. They have already got relationships that made sense because the people that they're serving are primarily the Robinsons. We, I gave you era one. We know what we did.
We're proud of it. We built a ton of skills, and now we're in a position where we can keep building on that. We will have to earn our way into some of those agencies. Previous to really the last few years, we really haven't had a great reason to go in. Certainly, we're in those agencies. I think we're well-respected in those agencies, but we're a niche player for certain situations that come up. We're not a mainstream player in the Robinsons serving agencies. A lot of what we talked about today is how we will get to that. It won't happen on Tuesday of next week, but I hope you take away from this a sense of, wow, this is a fairly clear strategy. They do get a lot of these customers. Can they keep them? Can they use those skills to keep adding and building?
Can they become, and I said it before, the contemporary solution for Robinsons of tomorrow? I think we can, and I think agents will absolutely listen to that story. It will be a slower build, just because of the nature of the competition and their history. It will be a slower build than what we will do in direct, but a lot of the benefit of having a direct channel is when we can show graphs, and you saw some of them today, where the opportunities have already shown through in direct. We will show those to our agents and say, "It can happen." What do I have?
I wanted to ask. Howard Gleicher at Aristotle Capital. I wanted to ask a question about basically your core direct auto business. You clearly have one fairly formidable competitor and others as well. Looking out 5 or 10 years, is there a way to garner additional share of that? Now that pie is growing, of course. Is there a way for Progressive to grow additional share of that pie? Or the way you price your model, the other things, are you just assuming that can't happen, particularly because of the competitor?
I would say quite the contrary. We acquire, and as Brian said, not direct for direct, but in total, we acquire a large number of customers, and we are certainly second in that direct channel. We do not want to have to keep acquiring them. Keeping them is a way to increase that market share. If we keep filling the funnel at the current rate, but not losing, I am not having to replace the two that we lost with two new ones just to stay even. I would actually suggest that in all fairness, GEICO is a tremendous presence in this industry, and they do a lot of things right. My gut tells me that Progressive and GEICO could easily be, and you pick it, 10% market share increase by the two of them in relatively straightforward terms. If I had to pick winners today, clearly I am biased.
I get that. I am paid to be biased. GEICO, Progressive combined represent 20%, I am rounding up a little, 20% of the market share. Yet the solutions they have for people, especially as both are adding these kinds of bundling opportunities, it could be a very different landscape 5, 10, and 15 years ahead. We think it will be, and that is what we want to play into.
Thank you.
Bill Van Arnam, Principal Global Investors. Warren Buffett has said something to the effect of he views advertising, particularly television advertising, as one of the best investments he can make for GEICO, and he would spend more if he could. Your approach seems a little less aggressive, more methodical. Can you just compare and contrast those two kind of philosophies? Also this chart on page 83 where you have you and your competitors on this graph. I'm not sure if it's advertising or not, but everyone seems to be moving to the right. You had GEICO moving to the left, and I didn't quite understand that. Could you just show me that one?
I've got it. Thank you. I got it. Let's take the first one. Why don't you talk about sort of our marginal cost and so on and so forth?
Sure. Just in terms of advertising spend, yes, GEICO does spend a lot on advertising. There's no doubt about that. Actually, as a percentage of premium for the direct business, we probably spend more currently. We actually do in that regard as a percentage of premium. Our philosophy there is we will spend based upon the yield. If we think we are getting the economics of appropriate demand quotes, conversion, and PLE, that'll help determine how much we can spend on advertising. If we are successful in extending these PLEs, which is basically a lot of the message of today, we will be able to spend more money on advertising. There's no doubt about that. That's one of the factors that goes into determining sort of how much we spend on advertising.
Achieving success there will lead us the ability to spend more on advertising, get more people to quote with us, et cetera. We also want to be able to spend more money, but we're not going to do it foolishly or indiscriminately. We base it upon the economics, the relative acquisition costs I referred to in my short talk, and that's the basis, and we measure ourselves against that. Have there been times when we've gone above allowable acquisition costs? Yes, but we're not going to be indiscriminate about it and spend wildly over that. I think the key for us is getting exceptional policy life expectancy, and we'll spend a lot more on advertising.
I think it'd probably be the dumbest thing in the world to sort of suggest you take on Warren Buffett into something like this. The fact is, they must appreciate the same as we appreciate, that at some incremental cost per sale, you stop doing it. You don't want to have your last sale costing you thousands of dollars. There is a yield curve for how much you spend, and there is a maximum to that. Frankly, let's be a little bit-- we can both afford to spend more. It just doesn't make sense necessarily to spend more. I think Brian's explanation is the right one, that we will when we can. With regard to the chart, now that I know what it is on 83.
That is a research chart. It's consumers' perceptions, and we ask them their perception of different carriers around the value structure. Think of Progressive in era one as a total value play. It was price, because they didn't know very much else about us. There's this issue of caring for the customers and being a sort of a market leader, those types of things. What you obviously want is some positioning, which is the orange dot, if I remember my colors correctly. The orange dot is sort of an optimal position. Where would be sort of, in our interpretation, the best combination of value for the consumer, but in concert with those customer-caring attributes. I hope it wasn't just a passing comment when I said we're now providing what we think is the highest quality claim service that we've ever experienced.
There were a couple of things on my Post-it note there that also said that the consumers are giving us some of the highest scores ever, and yet we're doing it at the lowest cost. It isn't necessarily that you have to spend more money to do that, but you have to have the whole psyche of the company really customer-centric and delivering great customer service. That's in our CRM organization and our claims organization. That's where we're trying to move that needle. It's not necessarily a big capital investment, it's not necessarily a pricing, but it's how we want to move. We don't want to have either/or. If I have either/or situations, grow or be profitable, I can handle that.
It's the combination that's really tricky, and this is one of those combinations that it's going to be hard to move to that optimal position, but we're very committed to doing that. We have to let the consumers see us a little bit differently than what we've projected ourselves to be, which is largely discount, do this, do that, get another discount. We're not going to move away from that, but we want to show there's more to the company. I will tell you, I admit these are low frequency relative to the millions of customers we have, but the kind of notes that I get, specifically from claim service, I know we're doing the right things.
I know we're moving in that direction, and we just have to get more and more of our customers to appreciate that and want to stay with us, because why go anywhere else?
It's just about 4:00.
Do we have time for one more question or we?
We can take one more.
One more question?
One more. Do you want to take the one on the screen?
All right. Okay.
Your business model delivers consistent earnings, yet your capital return approach is erratic. I can appreciate the feedback. Large buybacks special dividends, et cetera. Given the nature of your earnings, why hasn't Progressive adopted a more consistent capital return strategy?
You have the last question.
Great. I actually believe we have a pretty consistent strategy. We always retain capital in the business to support the business growth. We keep it for regulatory purposes. We keep additional capital for contingencies, bad things happening, hurricanes, investment portfolios, et cetera. Above that, we'll return it if we don't see a great need for it. Now, the form of that return may differ, and that does differ from time to time. Certainly, the one component that is a consistent measure right now is the variable dividend. It's a variable annual dividend. You can determine what it is. It's one-third of our after-tax underwriting profit multiplied by gain share factor. That's a formulaic. That's a return that we know will happen January or February of any given year.
The other return would be a combination of share repurchases and/or from time to time, as Glenn mentioned, special dividends. To start thinking on share repurchases is we look at that as an opportunity, but we won't buy at just any price. We look at it as some assessment of what's the current market valuation versus our own assessment of the intrinsic value. If we think it makes sense to be a buy, we'll buy shares, and you'll see it each and every month. After a point in time, if we have much more capital than we think we need, the special dividend comes into place. How the capital returned may change from time to time. I think the capital return philosophy is pretty consistent. Always have enough to support the business, invest in the business first.
If you have more than we need for both regulatory or contingent purposes, we'll return it. That's just what we think we should do, and I think we've been pretty consistent in doing that.
I think that is the question and the end of the questions. I'll finish where I started. Thank you for those who, especially our guests, that show interest in the company. We appreciate it. We think a lot about what makes sense to cover in these sessions, and we try not to sort of cover things that we think you already know, but give you a better sense of what's yet to come. We know there's a lot we don't get to cover, but we hope today we gave you a sense of a positioning for Progressive in the next era. Thanks.