Welcome to The Progressive Corporation's Investor Relations Conference Call. This conference call is also available via an audio webcast. Webcast participants will be able to listen only throughout the duration of the call. In addition, this conference is being recorded at the request of Progressive. If you have any objections, you may disconnect at this time. The company will not make detailed comments in addition to those provided in its quarterly report on Form 10-Q and letter to the shareholders, which have been posted to the company's website, and will use this conference call to respond to questions. Acting as moderator for the call will be Julia Hornack. At this time, I will turn the call over to Ms. Hornack.
Good morning. Welcome to Progressive's conference call. Participating on today's call are Glenn Renwick, our CEO, John Sauerland, our CFO, and Bill Cody, our Chief Investment Officer. Today's call is scheduled to last about an hour. As always, our discussions on this call may include forward-looking statements. These forward-looking statements are based on management's current expectations and are subject to many risks and uncertainties that could cause actual events and results to differ materially from those discussed during this call. Additional information concerning those risks and uncertainties is available in our 2014 annual report on Form 10-K and our first quarter 2015 quarterly report on Form 10-Q, where you will find discussions of the risk factors affecting our businesses, safe harbor statements relating to forward-looking statements, and other discussions of the risks, uncertainties, and other challenges we face.
These documents can be found via the investors page of our website, progressive.com. We are now ready to take our first question. Carolyn, we are now ready to take our first question.
Thank you. At this time, we are ready to begin the formal question and answer session. If you'd like to ask a question, you may press star one on your touchtone phone. You may press star two to withdraw your question. After pressing the star one to ask a question, you will be prompted to state your name and company to help with pronunciation. Please be advised that during this process, you will be momentarily blocked from the hearing of the live call. To allow the company to respond to as many callers as possible, you will be limited to one initial question and one follow-up question per request. If your telephone has a mute capability, we ask that you use this function during the time your question is being answered to minimize any background noise.
To the extent you have additional questions, you will need to place your name back in the queue by selecting star one on your telephone. Our first question is from Amit Kumar from Macquarie. Your line is open.
Thanks, good morning. My first question relates to obviously the focus on miles driven and the loss cost trends we are seeing. We are getting a lot of incoming questions from investors that it's only a matter of time before these frequency trends surface at Progressive too. Can you talk about why we are not seeing these trends at Progressive and maybe loop that in with the discussion on the risk curve as to where you write business?
Sure, Amit. I suspect you're more than referencing a few data points that have come out with regard to frequency.
Yes.
Let me just sort of start by saying no matter what frequencies are reported, they're all samples from an overall population. Even if we do frequency by channel, we're going to have slightly different numbers. I think what you're more likely focused on is what would be a fairly dramatic difference, let's suggest with Allstate having reported recently and Progressive. I'm not going to even attempt to try to explain those differences, but let me give you a fair amount of color as to where I see things going. We spend a reasonable amount of time looking at our frequency and severity, but I'll focus just on frequency for now because I think that's the guts of your question relative to PCI data.
While we don't lay on top of PCI directly, right now are a little more favorable, i.e., our frequency is a little lower, we're a lot more comparable to the PCI data than we were, for example, Allstate. That can happen. You sample any population, you're going to have different outcomes. It's not a question of what's right and wrong, but starting to explain mixed differences. Mixed differences can occur through urban concentrations, rural concentrations, different states. Recognize all of these data points really need to be broken down. You referenced miles driven. One of the advantages that Progressive has suggested to investors before is that we have an early look system at miles driven.
Again, it's a sample of a very large population, but with our Snapshot population being sizable, we're very comfortable that we're actually getting sort of on a daily, weekly, whatever we choose to examine basis, any real detection of miles driven changes. I'm not going to sort of become a free data source and give you the specifics, but I will tell you in July, there was an uptick in miles driven. We had seen an uptick all year, but it had been reasonably constant. Then in July, on a July-over-July basis, a more significant uptick. Again, what's more important there is what kind of miles are being driven, and we're analyzing that from a point of view. I'll just give you one break that we use, and that is miles or trips, single continuous trips over 15 miles.
We're actually seeing that become a higher percentage of miles driven. Again, frequency will track with the type of miles, and we're seeing longer trips, and there's a different frequency for longer trips than there are for shorter trips. We simply are observing our frequencies. We report them as they are. Obviously, very favorable right now. I don't know that you should assume that Progressive is anything other than a large player. Our frequencies may be reflective of the environment. They may change. Whenever we see another data point that's inconsistent with our own, clearly, it puts our attention to make sure that we're not perhaps overlooking a future trend to that data point. Clearly, what we measure is what we measure. The real issue is which direction is it going. We are seeing miles driven up. We are seeing a certain type of miles driven up.
We believe that that could result in higher frequencies, but so far, we're not betting on a dramatic change in the frequencies, and we're very comfortable with what we have. All of that should be translated into pricing. It's really not a matter of what the frequency is, it's whether or not you've priced for that frequency and priced for that severity. Right now we're very comfortable that we've actually previously seen opportunities to take our rates up, believing that frequencies would rise. We're in a very comfortable position as we see it right now. We have mechanisms inside of our rates whereby we keep those rates moving up a small amount, and hopefully, we will match frequency and severity for a period of time. When we simply get out of match, that's when we take more dramatic or quantum rate changes, 1%, 2%, whatever is necessary.
Did that get at your question?
Yes. That is actually very helpful. The only follow-up I have is, obviously, we've talked about Snapshot, and you talked about the insight it gives you. 2.5 million, I guess, is the ballpark number of Snapshot customers. When you talk about insight, how should we think about what is the % of that book which is sort of transmitting real-time data back to you?
That's a fair question, and I don't have a particularly good answer for you. That is the size of the book, but not all those customers have an active chip, because once they've actually gone through the first six months, we will actually have them give the chip back, and we'll recycle it. While you can always make the assessment that more data is better, there's also a cost to more data. It's really a question of the marginal cost versus the marginal return of new information. We're looking at something in the order of 1 million trips a day. Think about it as not so much how many vehicles, but what we're really getting from it. Cumulatively, we're well over 13 or probably approaching 15 billion miles driven. A million trips a day is a big number relative to sort of sample statistics.
We're very comfortable that we've got that inside look, and while it may not have been the primary driving force behind Snapshot, clearly it wasn't, it's an awfully good side effect.
Got it. Thanks for the answers, good luck for the future.
Thanks.
Thank you. Our next question or comment is from Michael Nannizzi from Goldman Sachs. Your line is open.
Thank you. Just following up on Amit's question there. I guess the one question I have is, you have two, both you and Allstate, I'm not asking you to comment on their trend, but you have two carriers that are big, that write across the spectrum and probably have some overlap in demographic. You almost have two, and you have miles driven rising. I understand your comment on PCI. We literally have two end points in terms of experience. With a large pool, I'm just trying to say, it doesn't seem intuitive to me that we could be seeing two dramatically different experiences. I get your point on rate, but rate is a top line. I'm sorry, rate is more a function of top line, whereas frequency should be more margin related.
Yeah.
Is there anything that we just, not being operators, is there something else that could explain two really, really large populations with vastly different experience?
It's a very fair question, Michael, I'm not in a position to try to even explain someone else's issues other than the fact that the good news is we all measure frequency. It's not a make-believe number. For the most part, we can assume that both numbers are right. Why they are so disparate from the same population, I don't have a great answer for you. I certainly can give you lots of pieces why they will never be the same. In 30 years of being in the business, I often have trouble reconciling between channels and other carriers. This is a pretty big difference, and I've not seen that on very many occasions. I don't have a good explanation. If I had to speculate, I would say mix is probably a good chunk of that.
I mentioned sort of the urban and rural kind of environment. I don't have the ability, don't think anyone does, to be able to dissect the mix of the business to see if it could be an explanatory variable.
That one's a tough one. I think the key thing for us, and maybe this is just part of my nature, is I get more paranoid when things are going well. You can be sure that if someone else is seeing something We'll be looking, it's not something that we're worried about missing. We know how many claims are coming in on a daily basis. We know how to count those claims, and if we see any sort of upticks, you can be sure we'll be adjusting for it.
Got it. Great.
We're not seeing it.
I'm sorry. You're not seeing it? Okay.
We're not seeing it right now.
Okay. Then I guess just switching on the homeowner side, can you talk about, you mentioned some points in your letter, but as you're getting agents on board, can you talk about what that entails in terms of expenses, commission differences, and also just the information and education to get agents to kind of understand the differentiated nature of this offering, not just versus what you were offering before or what others offer, but just from the fact that you're moving placement from being just a more of a pure auto writer to now a bundled provider and looking to kind of be in that space. Thanks.
Yep. Good. I'm glad you picked up on that because Platinum, think about our relationship with ASI in its current form of being one quarter in, we've gotten within that first 90 days, we're out in the marketplace with a product that basically is just us, not one or the other or two together. It's us together, which I find to be the most exciting thing that we've done in a long time. I said in there it'll be a limited distribution. I'll come back and comment on that. You talked about costs and startup. There will definitely be some startup costs, but most of those will be just a reallocation of costs that we have in the marketplace. We're not necessarily hiring new people or so on and so forth.
There'll be some incremental costs, but probably not ones that will flow through to any degree that you would see them. On the other hand, the commission costs, we absolutely are prepared to pay more commission, we'll do so for these Platinum agents if they provide us with the exact intent, the bundled product that we're looking for. That should reflect. It'll be a little hard to see. I'll be more than willing to take those sort of questions after another quarter, six months. You may have trouble seeing that flow through in the expense ratio because it'll be relatively small. Why are we doing limited distribution?
Frankly, we really want to get at a class, I tried to note this a little bit in my letter, a class of customer that for the most part, I'll separate here just not very often, but I will for now, Progressive and ASI, the more preferred customer that while we've always wanted them, we've always had a product, have not been the sweet spot for us in the agency channel. This gives us an opportunity to literally get a data set of those customers and understand if there is truly a different pure premium that they represent versus perhaps a more mass distributed product. Obviously, we're making a very, we think an intelligent and informed, but to be fair, imperfect at this point in time, determination that that pure premium will be lower and support a higher commission.
As we get more of that data, which we're excited to start having come in, we will be able to, A, confirm that, or, as I hope is possible, recognize that there's possibly even more there as we address this pure premium segment that we really haven't seen before, of both bundled and a class of customers that a group of agents have access to that we previously have not captured. A little bit up on commission costs, but that'll probably be very small in the aggregate. To help you with that, I'm more than willing to give you some data on that as we have future calls like this. Our internal expenses, while we're allocating sales resources and rollout expenses, I wouldn't worry about that too much.
Okay, great. Thank you.
Thank you. Our next question or comment is from Jay Gelb from Barclays. Your line is open.
Thanks. Good morning. With Allstate saying it's going to pull back on expenses, which includes advertising, my sense is that could result in a slower growth rate in policies in force for auto insurance for Allstate. Does that offer an opportunity for Progressive to expand its growth across the auto product overall and the direct channel in particular?
The answer to that has to be yes. Just we hope that we're a recipient of that. If you think about the percentage of new business that we get when people switch, we're obviously one of those considerations. Calibrating that, good news is we report monthly, you'll get to see. Over time, we will certainly internally look at the percentage of new business that are coming with proof of prior from Allstate, so on and so forth. We do that as a natural course of business anyway. We have seen, starting at about May, John, correct me if I'm wrong there, that the paid search from Allstate and even Esurance to a degree, has actually gone down very significantly. We track those sort of things.
Their comments of pulling back on advertising, certainly we've seen that in the paid search area, which is the easiest one for us to get a clean handle on, which is great from our perspective. Obviously, we're not doing that. In fact, our ad spend has been strengthened just a little bit. This is a time where as long as we're very comfortable with our rate level, we've talked about some of the macro trends of miles driven frequency. These are all the things that we spend That's basically what we do. We spend our time trying to make sure we've got those in balance. As long as we think we've got them in balance, this is actually a good time for us. I would also tell you, building on my prior comment of being paranoid, when you grow fast, the book changes.
While we're big, and no matter how much new business, it doesn't change dramatically, we've got to be very conscious of new business effects. To the extent that new business comes in with prior insurance, that is actually better for us than those that have no prior insurance. Is it an opportunity? Absolutely. We've got to make sure that we do our job and not worry too much about what others are doing. Right now, we're positioned in a really nice way. I feel great about the marketplace. I feel great about our pricing, and we'll take it when it's there.
Thanks, Glenn.
Thank you. Our next question or comment is from Ian Gutterman from Balyasny Asset Management. Your line is open.
Hi, thank you. Glenn, I just wanted to go into a little bit of the first half results and understand what's happening on the loss side a little bit. When I look at an accident year loss ratio, I strip out the CATs, I strip out the development. In the first quarter, you were up about three points year-over-year, in the second quarter, two points. If I back out my best guess at the mix of home, it was also close to about three points. Why do we see the accident year go up three points if loss trends have been reasonably within expectations?
Those numbers there seem right to me. We're priced to a level, recognize that we're also at a point where our accident year results would be comfortable. There's no question the calendar year results are better than the accident year results. That's why we publish so that you can actually get to those things. I, right now, given what we've just talked about, market conditions, our ability to attract new business, I am perfectly happy to be close to our target combined ratio when I can get a disproportionate growth to go along with that. You're right, but I'm not concerned about that, and I think we're priced to the right pure premium trends.
Okay. I guess I was looking at it-
May I add to that?
Yeah.
I'm sorry, this is John.
Yeah.
I would add that, yes, prior year development has been favorable. The good news there is we've talked a lot about frequency on this call already, but our severity trends have been good as well. Severity has come in below what we projected and obviously reserved for, and that allows us to show the favorable prior year development. The other component that I think you need to think about when you look at that total is the expense ratio. We target a budgetary loss ratio that's driven by our expense ratio, and we've been able to drive that down some. Think about maybe about a half a point there. You're just looking at the loss ratio. Combined ratio, obviously calendar year-over-year is down a bit.
Accident year, yes, up a bit. Very confident, as Glenn said, that we are pricing to our targets and are in a good place rate level-wise.
Okay. Because I guess what surprised me a little bit, John, was I thought normally I have to go double-check this, but I thought normally we would see in a period of a lot of releases, and especially when those releases are coming from the most recent accident year, releases would be driven by severity coming in better than expected, and if that was the case, that would probably help your picks for the current year, right? It just seems odd to me that the picks are going up at the same time development is getting better. I would have thought those would go in the same direction.
I'm not sure I understand what you mean by picks are going up.
What I'm saying is just, again, if you look at your overall 2015 loss ratios are basically flat year-over-year, but development's helped about 3 points. The change in development's been about a 3-point help year-to-date. On an accident year basis, you've deteriorated by that amount year-to-date, right? It seems like the 2015 accident year is being reserved to a higher loss ratio than the 2014 accident year. The reason the calendar year is the same is because of the reserve releases from 2014. Right?
The prior year development is actually a little less than 2 points to start with. I'm not sure where the 3 is coming from.
Again, I was doing the delta, that's okay. Okay. I'll move on. The other thing, Glenn.
Can I-
I'm sorry.
Let me just make another point here, and I'm not sure I'm directly on your question, but historically, we will react very quickly to trends because the penalty for not doing so is severe. Occasionally, we get out ahead of trend, and it doesn't materialize, or competitors don't necessarily take the rate increase. If you want to put it in the harshest terms, our penalty function for being out in front of trend that maybe doesn't materialize is wider margins. Here we think we've actually got our pricing very close to what our run rate accident year would need to be. To me, I don't think about a deterioration in loss ratio as necessarily a positive or a negative. I'm much more focused on the target that we're trying to achieve under the different market conditions.
If we get out ahead, and we've done that, we actually end up with wider margins. This is a time I don't think we're out ahead. I think we can bring in a good volume of new business and our renewals with it at our objective target. That deterioration seems like a negative. I would caution, at least for me, to say we're moving towards a more optimal ordered pair of pricing target for profitability and the ability to grow as fast as possible, which is exactly what we say is our primary objective.
No, that makes sense. Okay. Then just real quick on the, I guess this is as much a severity question as frequency, but I believe the national highway data showed an increase in vehicular deaths this year. In talking to some of the life insurance companies who've seen some adverse mortality, they've discussed an increase in auto-related deaths. Have you seen that pick up, and has that been an impact on severity at all?
You know what? I'm not really able to be informative on that one.
Okay.
You're probably right on all points. I don't have that information. I'd just be speculating, I don't like to do that.
Got it. Okay. Thank you.
Thank you. Our next question or comment is from Paul Newsome from Sandler O'Neill. Your line is open.
Could we talk a little bit about the favorable reserve development and the components thereof in the first half of the year? They do seem a little bit larger than what you historically had. If I recall, you folks are one of the better ones at trying to actually get kind of dead on over time. I mean, is there changes in the reserving philosophy? Is there anything particular that's kind of driving a little bit of an increase in the favorable development?
Why don't you take that with certainly the philosophy.
Okay.
There's no change. We've got Gary, our Chief Actuary, with us. Let's let John take that, and Gary, come in.
Sure. We have been quite accurate over the past several years when it comes to reserving, and it is always our objective of being as close as possible. There's been absolutely no change in reserving philosophy. The year-to-date experience has simply been that severity, especially for BI, has come in materially lower than we expect. If you look at a long-term trend line of BI severity, it's very positive. I think three, four points a year, we are down below that right now, and we think that is due to some improvements in our claims handling. We think it's a good problem to have. This doesn't mean that we'll dial everything down quickly to the new experience, but Gary is certainly watching that closely, and we'll adjust as appropriate.
We're seeing the improvement across business lines, that's another indication that there may be a handling improvement that we are seeing within our claims organization because our claims organization handles all lines within the same org. We feel great about the BI severity trends we've seen, and we'll take the adjustments as we see fit. On the property side, we have taken some adjustments there as well. It's been a little more in terms of percentage of losses than on the vehicle side. We're just beginning to really ensure the process around those reserves, and Gary is getting very engaged with folks who run ASI to ensure we have similarly designed processes around the ASI reserve. Anything you want to add to that, Gary?
No, I think that's a very good answer by John. The only thing I would add to that is, geographically, the development we're seeing is fairly consistent across most states as well. It's not just a few states driving the development that you're seeing and the improvement in the loss costs.
Gary, why don't you do the split commercial personal auto.
With respect to commercial versus personal auto, we are seeing fairly favorable development in both of those sides. On the commercial side, the development is more so in our business auto and contractors business, that's making up roughly about $40 million or so of the development that you're seeing this year.
Do you anticipate that as the property comes on, that we'll see a little bit different reserve pattern over time? My sense is that it's a little bit easier to be dead on an auto book than it is on a home book. Maybe I'm wrong with that.
I think it's a little early for us to project what we'll see there down the road. I think it's a little early to answer that in our case.
Keep asking it, Paul, our overall philosophy will be pretty much as you've known it for an awful long time. We're not interested in worrying about what gets reported in a calendar year. We live accident year, and as does ASI. We will always attempt to get that right because that's pricing in the marketplace, and we know what we have to price to.
Just one quick one if I can squeeze it in, that has nothing to do with the reserves. On the new product, Platinum, that increase in commission, is that a straight commission kind of increase or is that related to a contingent commission?
There's components of both. Agents should see this as quite favorable. I'll just hit some top-line issues. We won't go into all the details, but if I'm an agent producing, I will absolutely get an immediate recognition of more commission. Certainly from the Progressive side, it'll be additive to what I would have otherwise got. There is a level at which you can attain different compensation overrides later based on certain levels of production. While I know that, I couldn't explain them all to you right this second. We're very comfortable that that level of commission puts us in a competitive situation.
Not necessarily overly competitive, agents certainly will not be able to say, "Gee, I'd love to use you, but your commission isn't quite what I might get from another carrier." It's in that ballpark, those that actually use us very well can even expect more than that. We've made what I would call, not a concession, just a marketing approach. It'll be a very small percentage of our policies in force. We've also tried to coordinate policy periods so that those who are truly committed to that destination type of situation that we've been talking to you about can have an annual policy and a home policy with concurrent or simultaneous renewal dates. Those sorts of things are changes. We know the pluses and minuses of doing those sorts of things, but they've been very well received by agents so far.
Can't give you a great deal of insight as to Platinum's receptivity. We've been in the market in Texas about 40 days as of the end of the quarter. Indiana, I think, snuck in just in the quarter, if I remember correctly. I don't know if we wrote a policy in the quarter, but we're in there, we'll continue to roll that out. I will tell you that the initial receptivity from sales folks, now, sales folks tend to have a way of being more excited than sometimes I am. Is very, very positive. I'd be disappointed if that wasn't the case. I can't give you a really good feel right now about sort of incremental volume over the offerings that the agent already had. We are certainly of the belief that it's incremental.
We just don't have a good sizing on that that we would feel comfortable giving to you at this point in time.
Thank you very much for the answers, congratulations on the quarter.
Thank you. Our next question or comment is from Meyer Shields from KBW. Your line is open.
Thanks. Good morning. This is going to be a little bit long, but we're seeing an increase in average commercial policy size despite the fact that the results are phenomenal. I'm assuming that's mix shift. First, is that true? Second, if it is, does it drive your combined ratio closer to 96 faster than would've been if you just allow trend to erode the excess profits embedded in rates?
Only in a friendly way, Meyer. There's no such thing as excess profits. Yeah, there is absolutely some to that. Our commercial business in general, and we gave you this in the Q, is, average premiums are actually up, period, across the board. Is that on all segments? Yes. Is it more dramatic in some segments also? Yes. The mix shift towards for- hire business transportation and specialty lines with higher average premiums, absolutely, that's part of it. Your second part of the question was, should we expect erosion on the combined ratio because of that? The answer is not by design in any way, shape, or form. We try to price all pieces of our business. We call them CMTs. It doesn't really matter, just market tiers, commercial market tiers. We're looking for the same kind of margin in each segment.
While we've been able to get rate and have a good loss result is in large part reflective of the market really hardened in those segments, and there are less availability of carriers. We're benefiting from that. We're happy to do that. We don't think that'll last forever. You also use 96 in some cases. We don't price every segment to a 96. Our clear explanation in the Annual Report is the amalgam of all of our products comes to a 96. In places where we believe there are higher risks, and there are higher risks in commercial simply because you have lower unit numbers and lower losses to deal with. Our, what we will call indications, always have a little bit more volatility to them. We include volatility in part of our targets.
A lot in there, but no, you shouldn't assume that because we have a slight mix change to the or higher segments, which is driving average premium, that necessarily that should drive loss costs. We price each piece to be consistent with our targets.
Okay.
Yeah. I might add to that in the Q, we did tell you new business for commercial average premium there is up 20%-25%. That clearly is indicative of the shift that Glenn mentioned. We look very closely at the profitability by new business and by renewal business and by the marketing tiers that Glenn mentioned. We not only feel great about our profitability in aggregate, we feel great about new and renewal by all of those marketing tiers. We're watching that pretty closely. It is shifting for sure. We're feeling pretty good about what's going on. Just two other segments that we'll just comment on is that more recent, or let's just say later in the quarter, we started to see a little bit more active growth in our business auto, which is the higher number of units, but lower average premium business.
We're starting to see some growth in that. The first growth really came in the full higher segments. Not yet at all confirmed, but we're starting to see an uptick in the contractor business, and that may be a little bit more reflective of some declines and now just a little bit more recovery in the economy. At least we're starting to see the early signs. Don't think we'll be able to comment with any veracity until at least another quarter, but we're starting to see some early signs. The incoming quoting mix changed, but it's also starting to change again, and for us, that's a very good thing. A good balanced book is great.
Okay. That was very thorough. Thank you. I understand that you don't report comprehensive loss cost trends because there's a lot of volatility there related to the weather. If you look at other components of it, and I'm thinking here specifically of theft, is there any sort of macroeconomic trend? I don't know if people are doing better economically, there's a reduction in theft or anything like that.
Good question. I see Gary scrambling for some notes here, so I don't have an answer for you, but he may.
What I have in front of me is, in terms of total comprehensive trends, and this would be excluding CAT information, what we're seeing, severity is running around 3%-4%, frequencies relatively flat. I don't have the theft broken out right now, but that's easy enough to get if we want to get that information.
Gary was quoting the most recent quarter numbers there.
Correct.
Okay, great. I really appreciate it. Thank you so much.
Hopefully, the long-term trend on that will be very favorable, I mean long-term trend, pretty soon, cars will be able to tell you exactly where they are. Hopefully, we'll lose less of them.
All right. Good luck with that.
The criminals might get smarter. I don't know.
Thank you. Our next question or comment is from Brian Meredith from UBS. Your line is open.
Yes, thanks. I guess first question, can you give us an update on the success you're having or traction with Snapshot, the new Snapshot product in the agency segment?
Sure. Snapshot continues to be a big exciting venture for us. Probably the biggest things we've done, we gave you an update at the IR meeting. We've brought it into the marketplace. The terminology internally we use is 3.0. Don't worry too much about that. We put it in concert with the most recent overall personal lines product design. That's out now five, six states, something like that. In the four states that I have received data on and looked at, here are a couple of the points that are notable. The increase in usage of Snapshot by the agents has gone up fairly significantly, but still at a low level. All right? That's important. Absolutely, we achieved something that we were setting out to achieve. Did we get to a level that we would say approximates the direct? No.
Think of that as more in the 10% range, in terms of usage by agents. Whereas with regard to direct, think about it a little bit more in the 30%-40% range. We're not going to give absolute estimates on that, but I'll give you those ranges, and you can assume that they're very accurate. One of the things that is important to us in Snapshot is that if you take the customers that are, and specifically are now in direct, that are using a multi-product quoting mechanism, actually the numbers are considerably higher. Those who intend to buy multi-product from us are quoting and buying at rates closer to 45%-50% using Snapshot. We're actually very excited about that. A term that we may have used in the discussions in May is sort of the future Robinsons.
Obviously, we use Robinsons, but the future Robinsons, those Dianes that are transitioning, that group is showing a high propensity to consider Snapshot. When I say Dianes, it could be with a renters product, or it could be those that have become just early Robinsons. It's filling a niche that's very useful to us. When you start getting into those numbers, while we can continue to say there are parts of the population that are just still saying, "No, I don't want that level of intrusion in my vehicle," whatever their reasons are for that. We're not expecting 100% by any stretch of the imagination, but it's a pretty good part of the marketplace that we actually want and expect to be a big part of our destination era. There are those that are buying more traditionally non-standard, which we clearly continue to want to offer.
They're not always interested in having quite that level of active involvement in their insurance and may or may not keep it very long. Those that we want to sort of be with us at the best rate are actually showing very high take rates. Comfortable with that. I gave you some information on the mileage driven. In terms of our discounts, we're comfortable that at renewal we're seeing something around 60+% of people actually getting their renewal quote, which gives them at least the discount we advanced to them at point of sale. There's still another percentage that are actually getting a discount, but maybe not quite as much as we advanced at the point of sale.
That general 70/30 we've shown you in the past Is suggesting that there are still sort of 70%ish, possibly a little more, a little less, depending on the state, that are benefiting from the Snapshot. The greater good theory here is clearly working in our favor. There are a percent of those that are not exposed to a significant increase. Frankly, there are those that use a disproportionate amount of the loss costs, and we are applying a surcharge with this new model. We recognize that will be, for those individuals, either a non-renewal or a reluctant renewal, whatever that might be. As long as the numbers are as favorable as we see them, we're comfortable that we've got a good design out in the marketplace that can sustain and match loss costs better with the individual.
Great. Thank you. Next question, just curious, I see that you added two new states in the agency segment for the PHA program. How is that going to kind of play off against the Platinum product? Is there going to be both PHA and Platinum in state? Certain agents get just one than the other? How's that going to kind of work going forward?
You got it. That's essentially how it's going to work. There are agents that are very happy to have ASI Home and Progressive, but are not necessarily the agent that says, "This is a must-have. I'm going to build around this and put it as one of my primary providers." For those who are willing to make that commitment, we're willing to commit more, and that's what the Platinum program is all about. I will tell you that the interest from agents is meeting expectations at that level. Now I'm doing a personal commercial, but frankly, I wouldn't take the bet against a combination of ASI and Progressive. For the future, what we can do when we get the data together, I think we can really provide agents a serious product for their bundle higher-end customers. I think many are accepting that.
I'd add on to that commercial and say the first step is the renters for a lot of households. In the states in which we've rolled out renters so far, we appoint 75%-80ish% of our agents with the renters program so they can bundle auto and renters. We get better retention, when we have that package. We have a far greater likelihood of graduating them to the home and auto package as well. Obviously, that's with a little tighter distribution with PHA, yes, the pinnacle or Platinum set agencies an even tighter distribution.
Great. Thank you.
This is Julia. Thank you so much for everybody's questions today. We have great participation. Quite a few people still in queue, and we're running out of time. I'd just like to encourage people to limit their questions to one question and one follow-up, please. Thank you. Carolyn?
Thank you. Our next question or comment is from Ryan Tunis from Credit Suisse. Your line is open.
Hey, thanks. Good morning. Just a couple of quick ones, I think, and hopefully quick ones on the property business with ARX. During the quarter, net written premiums grew a little more than we would've expected. I'm wondering if there's any portion of that, we can attribute to higher retention or different reinsurance buying or anything along those lines.
No. Why don't we try to give you a little bit of insight to those numbers? I know they're confusing in some part. There'll be some changes in the way we address the direct written premium with reinsurance accounting and so on and so forth. Bottom line is, ARX's first quarter and second quarter were not dramatically different. While we didn't have ownership at the levels we have and consolidated reporting, we haven't given you that. If you really want to look, you can find the statutory numbers out there, and I'd say the second quarter is roughly the same as the first. Nothing in this league of dramatic that is worth reporting or a dramatic change in retention. Frankly, in the 90 days, while we're extremely happy with everything that's going on, there's nothing dramatic to report.
Okay. I guess just on the expense ratio, sorry.
Do you have something?
I would just reiterate what Glenn said. We are changing the way we account for our reinsurance obligations that take place normally around June and December each year. That net written premium may not be wholly indicative of the growth rate. The direct written premium will be much more indicative of the actual growth rate. I would focus on that number. We can't obviously report prior year, because we didn't have the ownership stake, so we can't give you apples to apples right now. Their second quarter statutory data will come out fairly soon, and you'll be able to see the direct written premium growth in those statements. I would also add, though, that while growth in homeowners is great, the more important growth is growth in bundles.
That was the intent of the transaction and come together far more closely than we previously did, and that's where we're going to get the biggest benefit. If we could overnight bundle the entire ARX book with auto, we'd be way happier than growing the homeowner piece. Just want to reiterate, that's the intent of what we're trying to do there.
Closely coupled with that, when we talk bundle, you should think policy life expectancy and amortization of fixed expenses and the like, and improvement typically in loss costs. Those are the things that really are as I think I wrote once before, the currency of this transaction.
Yeah, that's helpful. I guess just on the expense ratio, again, trying to reconcile that in ARX to some of the stat data we've seen. I think ex- amortization cost is sort of low 30s. Is that a good run rate to use going forward? Is there anything we should be thinking about there in terms of a jump up in expenses or anything of the like?
I'd make two comments on that. We do have Trevor Hill here, the CFO of ASI, here with us. First, I want to clarify a comment I made earlier in the call talking about our expense ratio being down a half a point. That is on the vehicle side or personal lines side. You add in the ARX expense ratio and look at our total, we're flatter. The ASI expense ratio, net of amortization expenses, is more in the low 20s. I think 24-ish. Correct me if I'm wrong on that. Trevor?
No, that's correct. It's between 23%-24%. We've held that for a few periods. I think one thing you mentioned was comparing our statutory numbers with the numbers reported in the Progressive numbers. Is that correct?
I did, yeah.
The difference there is mainly with our consolidated group that's reported with Progressive includes our managing general agents, which is not included in the statutory results. The statutory results are just insurance companies, so it won't reconcile one to one.
Okay, understood. Thank you.
Thank you. Our next question or comment is from Todd Bault from Citi. Your line is open.
Good morning, everyone. I wanted to ask a variant question on something you raised earlier about high growth rates could change mix. For what it's worth, our research has shown that over the last few years, you've done a pretty good job of your pricing and loss trend tracking each other within 2 points of course. Who knows if that's correct, but knowing what you know about your pricing and monitoring systems, if you went through a period where your pricing was more or less tracking your loss trends, would that have the ancillary effect of maintaining your portfolio homogeneity? In other words, your risks would be more or less as you expected, or are those quite independent? In other words, could you have pricing track loss trend well, but you could still have pretty large shifts in homogeneity. That's all I have. Thanks.
No, I think our attempt is to always have pricing track our losses, and to the extent that a new inflow of business changes the homogeneity of the book in place, it's always easier to manage a static book. It's not much fun, but it's easier. The real question is the new incoming mix equivalent to, similar to, different than what you currently have? We don't expect that to necessarily be the case. In fact, this is conjecture, so not a statement of fact, everything we are doing and seeing, and I've reported several of the trends today, Snapshot usage, how that's trending to multi-product customers, the notion of Platinum, we have every reason to believe the mix of business coming in may be more preferred overall than what we've seen before. That would be a positive.
Your pricing, unlikely to have any great effect, but should be positive to the loss cost.
Okay, thank you.
Thank you. Our next question or comment is from Al Ciamporcero from Columbia Management. Your line is open.
Thank you very much. I wanted to go back to a couple earlier questions just because I personally was a bit confused, and that's probably my own fault. In response to questions from Ian, you all spoke about the importance of looking at the overall calendar year combined ratio. In response to questions from Meyer, I believe you said that your focus is on the accident year more so than the calendar year. Number one, I want to make sure I understood that right. Maybe there are two separate issues being discussed. The secondarily and part of that was it does appear, as Ian was saying, that the reserve releases remain strong, but the underlying loss ratio continues to be higher.
I don't know if that should be cause for concern or if that should be cause for future optimism that the reserve releases may in fact be a leading indicator of an eventual improvement in the underlying loss ratio. I wonder if you could respond to those two issues.
Yeah, I'll try. I don't remember the specific reference to calendar, but that may be me. Let me just sort of suggest to you, as far as I'm concerned, we think in terms of accident year, most of the time. Calendar year is just sort of an adjustment for the quality defects and reserving in a previous way, in a previous period. Accident year for me is almost exclusively the way we like to look at the business. Clearly, we all understand that those adjustments are made from an accounting perspective, but accident year is probably a better way to look at the economics versus the accounting. I thought I tried to get at this with Ian's question. Maybe I wasn't clear. A degradation in the loss ratio under certain circumstances might be a really bad situation.
If, on the other hand, the loss ratio was lower than your previously designed target, it might be perfectly acceptable to move towards your target. Clearly not acceptable to go beyond your target. If and when market conditions are such that you're willing to give up a point of loss ratio and get what you consider to be proportionate volume or the elasticity of the market is willing to give it to you, those are trade-offs that not only do we make It's sort of the heart and soul of what we say is the guiding principles of Progressive, and that is we attempt to make a 4% underwriting profit. If and when we do, we try to grow as fast as possible.
It's very much that sequence, we are quite comfortable right now seeing that the market will give us some growth, we want to be priced as close to our target as we can. Hopefully that clarifies it, think accident year is by far more of the economics of the company. The calendar year is the accounting.
That's helpful. Thanks very much.
Thank you. Our next question comes from Joshua Shanker from Deutsche Bank. Your line is open.
Thank you for getting me in at the end. I appreciate always all the tools you give us to analyze things. I asked this question on the Allstate conference call, but not sure that I'm any smarter. Can you tell me the difference between paid frequency trends and incurred frequency trends? To the extent, is incurred frequency generally a more leading indicator?
I'm actually going to probably ask Gary to determine that. I will tell you that from my perspective, I spend most of my time on incurred. Paid may or may not be an indicator, not necessarily one way or another, sort of which way ultimately your prior trend picks might be adjusted. Gary spends more of his time on this. If I were you, I would focus more on incurred frequency.
Yeah. This is Gary S. Traicoff, Chief Actuary. I think just to add to Glenn's point, we look at both, I typically rely more on the incurred frequency as opposed to paid. Couple of issues you could have with paid frequency is when you have changes in closure rate, for example, that could distort what your paid frequency looks like, whereas the incurred frequency should be a lot more stable.
When you spoke about July frequency being up, is that a spike relative to July 2014 or relative to June 2015?
I think we should clarify the earlier comments. Glenn said vehicle miles traveled are up in July. He didn't say frequency was up in July.
Good point.
He also commented that the mix of types of trips has changed. I think of it as gas prices have gotten more competitive, more people are taking vacations instead of staycations this summer. We did not say the frequency is up in July. Miles traveled is up, and up more than it's been. I think we showed you back in May, it's been up for a number of months now, and July took a little further tick up.
Perfect answers. Thank you very much.
This will be our last question.
Thank you. Our last question comes from Vinay Misquith from Sterne Agee. Your line is open.
Hi, good morning. Just wanted to follow up on one thing, was the accident year combined ratio ex-CAT on the personal lines business. The first quarter year-over-year, my numbers show that the accident year combined was up around 2.9 points. Second quarter, that slows down to +0.6 points. Curious as to what's happening there. Why is the rate of deterioration slowing? Have you been taking pricing up or have loss cost trends slowed recently?
You have that one, John? Not sure what numbers you're looking at. You're saying accident year loss ratio through the first quarter of 2014 relative to the first quarter of 2015, and then second quarter 2014 versus second quarter 2015?
Yes. That's for only the personal lines. Correct. I'm talking about the combined ratio.
You're talking about the accident year combined ratio.
Yes.
Okay. Which the expense ratio will be the same. The loss ratio would be the driver of those.
Yes.
I'd go back to Glenn's answer on that issue. We are always pricing the business to, and I would add maybe a little confusion the way I said actually, and say, when we price business, we're pricing the lifetime combined ratios. There are instances where we are growing segments faster or slower, and the new business is always going to be a different combined ratio than the renewal business, and the mix of that business can drive calendar effects. We are pricing to the lifetime combined ratio by segment, but we're also managing the company to that 96 or better calendar year constantly. There's a couple different objective functions we are meeting there.
I think, again, we're really comfortable with our pricing across all business lines, and we think we are doing exactly what Glenn said, which is meeting or beating that 96, and we want to grow as fast as we can. I'd also encourage you to think about, at least in the case of Progressive, as pricing is more of a continuous variable than a discrete variable that happens every so often. We give you in the Q, the average premium, the rate of increase of average premium. You can see that's reasonably healthy, which is also going to affect the loss ratio as long as that average premium is going up faster than the rate of increase in the pure premium. Just to factor all those things in, there are a lot of moving parts. This is more a informational comment.
One of the things that we do is we track all rate revisions done on a monthly basis. That's usually for us that we track about 200-400 rate revisions. Think 50 states, many carriers. We're seeing on average, and if I cluster the last data points because it's quite a variable function. If I cluster the last data points, I'm seeing the market rate of increase at about 2.6, give or take. There's a give or take to that. We gave you an average premium increase of 4% in our personal lines business. We're at least sort of keeping up with that or perhaps our pricing got ahead of that, and others are now perhaps catching up. We also have some mechanisms in our rate in many states where we're actually continuously taking just a little bit.
It doesn't account for a great deal, but it means that each time we can soften the discrete event of having to adjust for rates. Don't be surprised that you'll see over time a little bit of closure, even in the situation that you're recognizing, other than our pricing targets, which is the biggest one, but you will see premiums change at a slightly different rate than pure premiums, and in some part, that's by design.
Okay. Thank you.
We find ourselves just past the top of the hour. Carolyn, I'm going to turn it back over to you for closing scripts.
That concludes The Progressive Corporation's Investor Relations Conference Call. An instant replay of the call will be available through Friday, August 21st by calling 1-866-470-7047 or can be accessed via the investor relations section of Progressive's website for the next year. Thank you for your participation. You may disconnect at this time.