前進保險 (PGR) 2012 Q4 法說會逐字稿

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 annual report on Form 10-K, annual report to shareholders and letter to 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 Matt Downing. At this time, I will turn the call over to Mr. Downing.

Thank you, Wendy. Good morning. Welcome to Progressive's conference call. Participating on today's call are Glenn Renwick, our CEO, and Brian Domeck, our CFO. Also on the line with us this morning is Bill Cody, our Chief Investment Officer. In addition, I'd like to introduce Gary Traicoff, our Chief Actuary. Gary assumed this position earlier this year when Al Neis, our previous Chief Actuary, retired. Gary will also be on the conference call today.

The 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 2012 annual report on Form 10-K, where you will find discussions of the risk factors affecting our businesses, Safe Harbor statement relating to forward-looking statements and other discussions of the risks, uncertainties and other challenges we face. Our documents can be found via the Investor's page of our website, progressive.com.

Wendy, we are now ready to take our first question.

(Operator Instructions) Mike Zaremski, Credit Suisse.

Good morning. Thanks. So the combined ratio over the past three months, in particular January, has been running well below 96%, which, Glenn, you seem to have reiterated in the letter to shareholders as an important target level.

So I'm curious then whether forces which have caused the combined ratio to run well below 96% are more temporary in nature. I guess, for example, perhaps there's an element of, quote, unquote, corrective actions following the combined being above 96% in the first half. I know you guys have talked about expenses as well. Thanks, and I have one follow-up.

Okay. Fair question. Our rates clearly -- I try to outline that fairly extensively in some of the communications -- but we clearly corrected our rates in mid-year. Some of that is earning through and earning through at levels that look like will be well suited to meet our 96% or below.

I wouldn't personally read too much into like a January number. There were -- winter months are interesting at best, and in some cases, we had some winter weather that was a little more typical and some that was quite atypical. We even had events that probably took out a -- for a large part of the population -- almost a weekend's worth of driving.

So we don't know and nor will we ever be able to sort of get to the point of being able to say that was a half a point or a quarter of a point or even a couple of points of CR.

We think we are priced at about the right level for going forward. As the nature of rate revisions, you tend to be a little bit spiky. You take every revision, and by definition, you are taking those rates up to a level that is slightly higher than you might want to if you could bleed the rate level in over time. So I don't want to get too technical here, but we price to a midpoint of the rate revision. And so in the early phases of a rate change, we are likely to see a little bit more of an earned premium. As the earned premium comes in, we are likely to see (inaudible) perhaps reflect a slightly better combined ratio. But over the length of that rate revision, we would expect to meet our 96% or below target.

So I would tell you right on track. But think of the maturing of a rate change. So we take it up a little bit, it matures through, and we are now coming into -- and I don't want to overplay this, because the results will be the results -- but we are starting to come into little bit more of a mature place in our rate revision. I'd add a couple of points to that.

So it means that people who are about to renew are now going to be renewing into the rate revision they were last on. And that's an important point, because they don't see the quantum jump in rates, which they see the first time after we've taken rate changes. That's sort of what I mean by the maturing of the rate level.

We also would expect to see from the time we've taken our rates competitive actions that start to make the competitive nature of the rate change a little more favorable to conversion. Does that get at your question?

Yes, that's very helpful. Lastly, I noticed that policyholder life expectancy levels for the full year 2012 were well below the levels for the first three months of 2012. So I think that implies there was a sharp dropoff in 4Q. If that is correct, what were the drivers, and what are the implications for the coming quarter? I guess, for example, does that translate into perhaps a sharper PIF decline? Thanks.

Okay. Actually, Brian had something to add, I think, to the past question, and probably will carry right on to this one.

Sure, Mike. The only thing I was going to add to Glenn's comment for your first question, a lot of it is sort of the earning in of rate change that improves sort of the loss ratio.

The other thing I would say, which we referenced in the letter and the like, particularly towards the last half of that last year, we did reduce our advertising spend, which was influencing the expense ratio. We have since the beginning of the year come up to more comparable levels to a year ago. So in January, we did ramp up our advertising more so than what we had in the second half of the year, since we were more confident in our rate level.

And the other thing, which we've talked about before, at least in the aggregate, when you look at the aggregate combined ratio, keep in mind some of our special lines products. There is some seasonality in terms of the losses for our specialized products; for like example, boats and motorcycles, much less a usage in sort of the winter months.

But for the auto perspective, most of it is the rate change related.

In terms of your question related to policy life expectancy, a large part of it and the decline is due to the rate changes we took. And we know every time we measure it in terms of retention rates, renewal rates, et cetera, when we take great changes there is a negative effect on that. What I would say is most of the rate changes and most of the significant rate changes we took were more in the late second-quarter, early third-quarter timeframe. Not all of them, but most. And most of those, as Glenn has mentioned, have cycled through the system, such that subsequent renewals will not see as big a rate change as they might have seen in the last several months.

So can't clearly predict what the PLE change will be, but certainly the decrease was a reflection of the rate changes that we made.

But am I correct in saying there was a big drop-off in 4Q, though, even though you started raising rates in 2Q and 3Q, like you just said?

Yes, the retention rate -- since policies for auto are six-month policies, each and every month, a cohort of policies are renewing. And so each and every month, yes, we have seen lower renewal rates than we might have previously seen. So that would be contributing to the deceleration of the policy in force growth. As well as certainly the rate changes have influenced our new business conversion rates.

Just to be clear about that, some customers might see the rate change the day after we make them, and some may see them six months after we make them; so it takes a full six-month cycle. Actually it's even longer than that, because there is a little bit of a delay from when we take them to when they are effective for renewals, because we quote renewals quite a way in advance. So we are not getting more detail on that cycle -- I would just tell you a cycle now has gone all the way through.

Thank you.

Josh Stirling, Sanford Bernstein.

Good morning. Thank you for taking the call, and Glenn, thank you for the letter. And Gary, congratulations on the promotion.

The question I'd like to ask would be -- it is obvious we were talking about the BOEs coming in, conversion rates coming down from the cyclical rate-taking. When you look across larger companies, there also seem to be shrinking units while they are raising pricing and people are still talking about severity. So the first part of this is really just do you still think severity is systemic, and this is an issue that's going to work its way through the rest of the system?

And then the question is, when we think about some of the lagging competitors -- regionals, mutual, perhaps, and not asking by name -- but broadly, is this something that you think will lead to an increase in the rate base at those firms over the next six to 12 months, or is this going to be a slower story to play out?

Let me see if I can get to some of those pieces. Clearly, we were -- I think we were pretty clear about what we saw and how we discussed our need for rate last year. So the real question sort of comes what do we see now.

Let's break it down pretty simply. Frequency, I would tell you not much of a story for frequency. It is fairly benign, and if you put a range around not much happening to plus or minus two, you probably capture everything we know about frequency by almost every coverage, and you are probably even closer than the two -- plus or minus two. That's good, but it can change at any point in time, so, as you know, we watch that very closely.

Severities, you will see from our most recent publication sort of five is a number that you can reasonably hang onto. We have some belief that maybe in the bodily injury range there might be some reason to believe it is a little less than 5, 4, 5; it gets really hard to sort of split hairs on that. What is driving it for us, it seems to be the litigated and non-litigated, but attorney represented, soft tissue claims. And there, we are seeing just a little bit of a difference in the general damages or the settlement ratio relative to special.

So we sort of have a sense of what we're looking for. Sometimes that can be hard to find. We kind of have our eye on the right things. But I think your takeaway, think about severity trend, or inflation if you like, sort of four to five. And you can put that on collision, you can put that on PD, and you can put that on BI. And I think that's how we see the world for right now.

To the extent that we are the size we are and have a representative book of business, it would be very unlikely -- although no one ever seems to match perfectly with your estimates of frequency and severity, we are generally all within a relatively close range. If others are experiencing the same sorts of things we're experiencing -- and I would just say it is hard for me to see why they would not be -- then it's over to them to decide what their objective function is. But we all know where interest rates are. It would be very hard to sort of think you could openly subsidize a lot of trend in your underlying book of business. And how and when they take rate is sort of their business.

I don't think it is hard to trace the breadcrumbs of the last four or five months. You've seen quite a few companies make announcements about their relative rate level. We track rate level, pretty much a scatter plot of every rate revision we can see, for every competitor, and it is clear to us that rates are generally on the rise. But in some cases, more dramatic; in other cases, not as dramatic.

We are looking closely at all results that are being reported, and we don't know what others will do. But we would be reasonably expecting there are still some rate changes to be taken in the marketplace if people are acting on the same data that we think we are seeing.

That's great. That's helpful. Thank you.

The other thing I would love to just briefly touch on is if you guys can give us an update on sort of your general thinking on Snapshot, both obviously iterating the marketing messages -- you've learned things over the past year in terms of both how you think you can sort of better position this vis-a-vis customers. It would be great to get some insight on that.

And I think we would also be interested just generally in sort of what your current thinking is on licensing and what kind of response you've had from companies now that you've been positioning your offers to the broader market for a couple months now. Thank you.

Josh, could you just quickly give me your first part of the question again?

Yes, most specifically, I would love to get a sense of what you think you've learned about consumer interest in the product and how you're going to iterate -- obviously, you're keying up a new campaign that we are all dying to see. We would love to understand sort of some of the thought that goes behind it and how you think the message and media evolves.

I think I try to put some -- for me at least -- if I say things like some of the most exciting things I've seen, that is like -- that is about as wild as I get in my commentary. So Snapshot is truly amazing for a lot of reasons. One is it is a lot more consumer friendly because it is a little bit more causal. They relate to that, than of the variables we use which are a little more correlated. So no issues there in terms of acceptance.

We get about -- and these are going to be numbers that you can reasonably rely on, but they will be not necessarily perfect at any point in time to the changes -- about 35% of our new business in direct is actually opting to take that. So it is a good percentage. The number is closer to a 10 in our agency distribution. But of very recent times, very recent being this year, we have actually done again one more push to our agents to see if we can get them even more excited about this proposition. And of recent times, we have had somewhere in the range -- and I don't know the exact number here, but my range will be good enough for this conversation -- about 8,000 of our agents actually take us up on a test drive. Which for us is very exciting, because when they've actually got more of a tactile feel for the product and what it does and how to install it and so on and so forth, we think they are more likely to be confident to be able to push that notion on to their customers.

Loss results continue to be -- the data is just so rich. We do sense that we will have product modifications. I forecast that even a year ago, that we're going to have to sort of figure out how to bring this into the fold of the greater product. So heretofore, it has largely been a discount, and we're going to have to sort of integrate that with the product. I alluded to that in my letter.

So lots of work to do, but how do we feel about snapshot? It is here to stay. There is no question about that. And how to use it. There are some drivers, and not a small number, that really do get a significant benefit. And as a segmentation variable, it is just clear it is just so powerful.

I mentioned in the letter, so just to try to anticipate things that might be on your mind, that test drive didn't meet expectations. That is true. And it might have been our expectations that were wrong. I will tell you that those that have taken us up on test drive, we do get at least some of the other statistics that we wanted to understand a little better. Of those that take a test drive, will they actually put it in? Will they do the whole test drive? Will they actually quote after the test drive. And is there some sort of proxy even in their taking it that tends to suggest that they may be the people that are eligible for a fairly significant discount. We've actually got some pretty good data on that. So while the numbers are not as great as we would like, the calibration and the metrics now are actually quite exciting for us.

And yes, we do have some new advertising, and probably if we had to critique ourselves, what we believe is that a lot of consumers sort of understand the notion. I gave you some statistics in my letter. They attribute it with Progressive, but they haven't quite yet figured out why are you talking to me. And we will try to do a significantly better job, and it will be a little bit more aggressive job in some advertising that will come out early April timeframe, where we will try to make it a little clearer for consumers. Actually, I'll take that back. We're going to try to make a lot clearer for consumers sort of the difference between if you are a good driver and you are not, the implications to your rates. So it will be pretty much a little more in-your-face advertising in that regard, and we hope that people will get the clear notion that Snapshot is a solution to a real problem and they will need to determine whether that's a problem they want to solve.

I think your follow-on was with regard to licensing. We are being clear starting mid last year that we intended to license this. We made our commitment to put out some of the details of licensing before the end of the year. We've done that. And I'm just going to sort of probably end there. I will say that we are very conscious of many efforts going on that are not necessarily under the terms of our licensing agreement, and those are best left for us to deal with at this point.

Okay. Thank you, Glenn. Good luck. It sounds exciting.

Vinay Misquith, Evercore.

Good morning. The first question is how do you plan to use pricing as a tool for generating growth? Do you plan to use -- do you plan to reduce price, or are you waiting more for competitors to take rates up in order to generate growth?

I'd tell you that we price to our costs. (technical difficulty) it is that very important combination of growth and combined ratio. And I tried to take a little more time this year in the annual report letter to say that's the combination, and frankly, that is the only acceptable combination that works for us.

So let me try to be specific to your question. We took rate because we felt we needed to and that best represented our view of current and future costs. We're happy with the rate. There are places almost inevitably where we get it a little bit off, and little bit over, sometimes a little bit under, and there are some places that will take some adjustment. But it's less about specifically the competitive environment. It is more a reflection of whether or not we are matching price with our cost.

So we have actually taken a slight decrease in Florida and in Texas. And these -- Florida specifically was a very tough state last year with things that were quite one-off situations. PIP was going in a very different direction. We saw new legislation in PIP. So there were some changes we needed to make in Florida. A lot of moving parts there. And to get that exactly right would have been more than heroic, and we will fine-tune that. That's what we do. We fine-tune.

Second is how do we get growth? We continue to really create a high demand function. While I don't want to get too far out ahead of the results that we have published, I would tell you that our demand, specifically on the direct side, is actually very strong. So the first half of the year or the first two months of the year now, very strong demand.

Our conversion, because of rate level, is not what it had been at the most optimal point of rate competitiveness in the marketplace. We sense that we are coming closer to that sweet spot, and it is a function of the age of our own rate revisions coming closer to the mid-point pricing and some competitive action and hopefully some yet to come.

So in short, I'd answer we generate demand and we are doing really well at that. I'm actually very happy with the demand functions, specifically on the direct side. Agency, there is a little more tardiness to the equivalent numbers from last year, but we're in the right ballpark there. So demand, I'm happy with. The maturing of the rate revision, I'm happy with. It may be yet another month or so before we are in an absolute sweet spot. We'll never know that until after the fact.

And competitive actions, everything I'm seeing in the marketplace would suggest that the rate will come to us, and if our conversion rate goes up a tick or two, we would be in a very advantaged position.

Okay, that's helpful. Just as a follow-up to that, on the conversion rate, historically, I've noted that Progressive is very smart. You drop pricing only if you can grow PIF. What if we actually don't see the conversion rates improve? Would you be willing to let the combined ratio slip more towards the bottom end of the range?

Say what you mean by bottom end of the range -- higher?

Closer to 90% versus 96%. So let's say that you don't see -- so we don't see the conversion rate improving; would you then say that it's a more competitive environment and we would rather go for profitability versus growth?

You know, I'm just not going to go there. We have a great product, we have great marketing and very clear objectives. We want to grow profitably.

So I know what you're asking, and that just isn't really on the table right now. There's no reason to believe that we don't have a product that the consumers want. And as our price point comes in, we think that -- well, they are asking for it; I know they are asking for it already -- while I would love to have a conversion rate a couple ticks higher than we currently have, my bet is that we'll see that a lot more than I'll have to take an action to say let's just in cheap terms sort of eke out more profit. I'm here for a long run with consumers, and I don't want to take that kind of action. I'm here for a long run with them, and we've said our preferred form of growth is new policyholders. So I wouldn't be banking on that.

Okay. Thank you.

Michael Nannizzi, Goldman Sachs.

Thank you very much. I guess just to follow up on that a little bit, Glenn. So it sounds like -- my initial question was going to be, if you want to grow, but you are not advertising, how are you going to grow? But it sounds like what you're saying is that there is like tide level of rate in the industry that is not quite where you feel like it needs to be. So interest is high, but the rate level in the industry is just not quite at that point where you will start getting that sort of conversion.

I guess I only question would be what happens if that doesn't happen right away, or this year or in the next six months? Are you willing to just continue to let PIF recede until the market kind of clicks with where you think it needs to be? And then I just have one follow-up. Thanks.

First of all, we are not pulling back on advertising. My letter, to the extent that I clearly indicated that in the second half of last year -- and I think your analogy of a tide tells me you are getting this point -- we pulled back in the second half of last year. But as I said, there was no good reason as we were seeing entering this year to get off of a run rate that's more comparable to the first part of last year. So advertising, we are back on the gas on that, and we have a very volatile valid product and a valid conversion.

We are not -- this is not like we are not converting a lot of people. We are making a lot of sales. A tick in conversion means a lot. I'm betting that that will come back to us from all the reasons that I've described. So that is -- to ask the question how long would we sustain that is sort of an impossible one to answer. We're always looking to refine things in the marketplace, even at the time -- this is a little more detail than you probably want it -- even the time that we took rates up, when we know we have customers that are coming to us that are not going to meet our profit targets, we also take some other underwriting restrictions. They might be bill plan type restrictions. We may have some early filters that we apply. And as we get rate, we are also able to go back and evaluate those filters.

So there is actually a lot more going on than just rate that can help us grow. And I don't see any signs, as I'm seeing data coming in, that would suggest that we are on the wrong track for getting our units back where we would like them in a reasonable timeframe.

Okay. I guess in thinking about -- and I know you talk about inputs, which makes sense. And we unfortunately kind of focus on the outputs on the earnings side because that's kind of what we see. But if we are here and you're kind of looking at this 96% combined, and you reiterated that as kind of an underlying component of your operations, and this sort of tide notion in the industry is something you can't really predict, and the portfolio stays where it is, how should we think about what allows earnings to kind of make that gesture back to like 2007 levels, if that is something that you think about, or maybe it's not.

No, and there's one other point too, as I broke off -- I'll come back to that -- but one other point that I broke off. It is an important point, so I'm going to reiterate it. Renewals are a big, big, big chunk of what we do. And people renewing into the rate level that they were previously on-- that was the point I made to an earlier question -- that becomes a very important point. So that would just be your first point.

Earnings, look, there is not a day goes by that anyone here is not focused on all the same kinds of things that you care about. We care about the same things. We are just trying to be abundantly clear of saying how we will get our earnings. And having a very clear objective function, which is why I took the time in the letter to reiterate that -- there was no new information there; that was a clear explanation of what we already do -- we are every day focused on trying to grow our book of business. We want to be a growth company, grow our customers.

We have an absolute threshold constraint of a 96% combined ratio that is not something we will violate. And that will be the driver of the operating part of the business. As we will support that with another huge part of our earnings stream in investments, I think we've been very clear about our philosophy in investments. So while instantaneously, we may be in a low interest rate environment, we think operationally, we are in a very good place. And while I kind of hate coming into the year without the momentum that we came into last year, my hope is that we don't have the inflection point this year that we had last year and that momentum will build from here on out.

So we are going to continue to be a very strong operating company, grow on the basis that we have given you, and when our opportunities for investment income are even stronger than they are now -- and I think we had a pretty strong year last year -- that will be an absolute delightful add-on.

Very much appreciate the answers. Thanks so much, Glenn.

Paul Newsome, Sandler O'Neill.

Good morning. I wanted to follow up a little bit on the investment question, in that how much did you re-examine or think about changing the investment strategy, given the low interest rate environment?

And I guess sort of as a second question, somewhat related, would that change if we have a different type of reporting? We are looking at financial instruments forcing the equity portfolio results on a mark-to-market through your income statement. And would that change how you think about your investment results? So two parts to that question.

Bill, why don't you take the investment strategy part of that, and --?

Sure. We think about it every day. And we are always looking for ways to improve our total return of the portfolio, not necessarily just our book yield or our GAAP yield, because we do run it on a total term basis with our goals of protecting the capital of the Company to protect our underwriting business and [enter] as much as we can. So it is always that balance of are we getting paid to take some of the risk that you need to take to improve your returns.

So we could easily increase our yield or our investment income by moving out the credit curve or moving down in credit or out the yield curve a bit. But to us, our judgment is that's not the best way to reach our long-term goals of boosting our total return. We are very mindful of the fact that even for a small increase in rates, whether its treasury rates or the spreads on non-treasury products, at the current low yield levels, that will produce a negative total return pretty quickly.

So we constantly evaluate it, but our philosophy and our goals always stay the same. What Glenn I think was referring to earlier is if the environment changes -- and as I think we all know now, the environment is pretty tough, with very, very, very low yields -- but if that environment changes and there were more opportunities for us, we would take a bit more risk. Our duration is at the short end of the range and it has been there for a while; again, to protect capital, not to take much interest rate risk. And our credit quality is high. And feel good about how we are positioned.

I've love it if rates were higher and we saw more opportunities, but that's not the case. So we are not going to stretch and try to hit some artificial yield bogey or some number.

And Paul, this is Brian. On the second half of your question, no, I don't think changes in accounting would change our investment strategy at all. Bill mentioned our objective function is on a total return basis. That's not going to change. The fluctuations in the equity markets there are reflected in comprehensive income, which are in the income statement. We had previously recorded on comprehensive income well before the change this year that have more reflected on the income statement.

And yes, if it all flows through the income statement, it will create a little bit more volatility in earnings per share, absolutely. But on a comprehensive income basis, which we think is a better measure of all-in, it is no change.

Thanks.

Meyer Shields, KBW.

Thanks. Good morning. Glenn, in past investor days, you've talked about different customer segments that have different retention tendencies. And I was wondering if you could talk about how the different segments' retention rates were impacted by the rate increases you took in the middle of last year.

Sure, I can do that. I'm not going to do it with great specificity, because it might be something others want to know as well. But unfortunately, it probably was across all what we call our CMTs -- don't worry about it -- just call it customer segments. And to the extent that if there was any slight bias, it would be towards the upper end of the client that we actually would like the most and the ones that stay the longest. So again, it is a bias that's there in the data. It is not sort of jumping off cliff kind of concern. But our more preferred customers clearly are showing that they don't like the rate volatility. I don't think anyone likes rate volatility, but sort of across the spectrum, with a slight bias towards the preferreds.

I would also tell you, just to sort of at least give a nice piece of good news there, I remarked on demand and relative comfort with demand as we see it through the first two months of this year. And I'm doing that relative to prior years for the first two months, because this was a high period anyway. We are actually seeing a slight skew to more preferred shoppers in our direct channel, as well.

We also put out -- and I commented on this -- our mobile applications; we've really quite enriched our mobile applications. And while putting three by three, which is three vehicles, three driver capability on mobile, it has also given a notable shift towards -- it's not huge numbers -- but a notable shift towards more preferred customers quoting on mobile devices, which is great.

On a personal note, it would be very hard for me to do a three by three on my phone, but that is probably more eyesight-related than anything else. But it is very clear that people -- and we are getting now an absolutely meaningful percent of our shopping coming in from mobile devices, whether it be iPad, iPhone or Android devices.

Great. Thanks. When you talked in the letter and earlier during this call about the free trial falling short, was that just in the number of people that wanted to use the free trial, or were there different steps in the process to conversion that also did not meet expectations?

No, mostly the top of the funnel. And it is worth repeating, because this is sort of important, because we are going to take another shot at this. Really, the top of the funnel, and that is why I said maybe our shortfall was really on making this real for other people, like compelling enough reason for me to go out of my way to try it. Insurance isn't necessarily the most engaging topic, period. So for people that actually engage in an insurance-related activity when they are not really highly motivated to do so, that puts a fair onus on us to make them move.

But of those that came through the funnel, and we did have a priori estimates of how many would actually install the device -- remarkably, people do take this and then never actually install the device, and that's important -- then complete the full trial period, then actually get a quote after the fact, we had estimates of all those things. Obviously, when it's brand-new, you have nothing but an estimate.

But in many cases, we were very close; in some cases we were able to show that we were conservative, and in others, a couple under, but in aggregate, pretty nice. So we're not worried too much about once you are in the funnel the kind of dilution effect that gets you down to the point of taking a quote and ultimately realizing a discount.

That -- the funnel looks good. Now we just need to fill the funnel at the top with a lot more consumer demand. And that's going to be tricky. It is just not something that is buying a Coke. It is very different, and we are going to have to make it very compelling for people.

Okay. Thank you very much.

Josh Shanker, Deutsche Bank.

Thank you very much. I have two questions, one very short-term and one very long-term. On the short-term, at the risk of sounding foolish, January was, as far as I can tell, the first January where you lost customers. Can we quantify whether the ad spend was high in January or low, the degree to which the tide had receded over the rate filings? And I -- just making such a leap that January is the [wrong] month, let's look at February and forward.

Advertising in January was -- it turned out the actual amount spent was a little less than the January last year. But that is more an actual versus sort of budget versus a plan to do something. There was probably some media purchasing that would have happened prior to that, so don't read too much into that. We are back roughly at the spend levels of January. And our realization of interest, I think I've already said, is pretty strong. And you can take that through January and even stronger into February. But let's not talk too much about February at this point; results will be out some -- couple of weeks.

So it is mostly -- you're right, PIF falling in January. No one is happy about that. But that is really just the same story that we've had several times in this call. You've got the rate revisions coming through, you've got people renewing into it, and for the most part, we are now through that cycle.

That would be the biggest reason. Brian, anything else to add on that?

No. Advertising, as we mentioned before, we ramped up to higher levels starting in January, ended up the media spend being a little bit less than last January, but still at pretty high levels. And more of the conversion. Conversion influenced that, depressed new business production, and a little bit on the retention side.

Okay, thank you. And on the very long-term, if five years ago you had asked me how much of the market would be Net driven, I would think it would be very high. Today, GEICO is probably about 8%, you're about 4%. I assume the whole industry is less than 20%. That kind of surprises me in retrospect. Is the market getting saturated? Why isn't there a higher percentage of traffic for buying auto insurance going through the Internet right now?

Probably a fair question. I don't know that we view it quite the same way. Of our direct book of business, an extraordinarily high percentage is Internet. So your numbers, I know where you're coming from on those. But we are probably very oriented to having our customers deal with us online. We believe that is a sweet spot for us; external reviews of websites and so on and so forth believe that is a sweet spot for us.

So of our direct business -- and you know the mix there and you see the vectors of change over time, and certainly they are a little less dramatic now than they were when you introduce something -- but call it 55/45, a little in the favor of the agency channels, a split that works. So even if you say 50/50, what you probably should take away for you say short-term and long-term, consumer behavior has switched pretty dramatically in a fairly short period of time. But you might conclude that it is stabilizing, in terms of how they want to buy, with an intermediary or without an intermediary. You might conclude that.

I don't think there is any need for us to conclude anything. We just want to make the options available and be able to be price-indifferent between the channels. But of the direct channel, a very significant percentage is Internet-driven for us. So you can sort of do your own math on that 50% and significant percentage.

And I also just alluded to the fact that mobile devices are a non-trivial -- I should give you some indication -- let's say more than 15% but not 25% of our direct business. So that is a pretty big chunk. And of the people that shop or quote on their mobile device, about 60% of those are actually then consummating the buy on the mobile device. The remainder, it's phone or they go to the Internet. And of that percentage, actually excuse a little bit more until they pick up the phone.

So you've got some really interesting behaviors. I think it is probably a little too simple of a segmentation now to say it's Internet or intermediated. The Internet can be a combination of a lot of different things. We actually have people start on the phone and finish on the Internet; start on the Internet, finish on the phone; start on a mobile device, finish on the phone; and lots of different combinations.

But frankly, we are largely orienting ourselves to be indifferent but highly, highly attractive for people who want to use some form of technology to get their insurance quote and sale. And I don't think long-term -- you say long-term five years. You go out even further, it is hard for me to imagine even in the agency world that that is not the trend that will dominate.

I think agents themselves are also figuring out how to use the Internet to build their own business. So it is not just, oh, it's only the direct channel that's using the Internet to market to customers. So whether it is, hey, what is the shift to the direct channel per se, I'm still pretty confident that lots of shopping will occur in the direct channel. But agents as well are using the Internet to their advantage, and we try to help support them in that regard.

I guess maybe it's the same answer, but is there any reason for me to believe that direct Net sales are being saturated?

I don't think so.

Okay. Thank you.

Ian Gutterman, Adage Capital.

Hi, guys. My first question is on the January results. I think historically in January, we tend to see a trend where there is very high favorable development and a very high accident year. And I think the interplay with that, as I recall, is December claims that settle in January and get recorded as development.

This January, we saw a little bit of the opposite. You actually have some adverse development and a very favorable accident year compared to historic January. So I guess I'm wondering, one, is -- are those two related? Is somewhat the better underlying accident year in January related to the lack of favorable development? And I guess the second part is sort of why didn't we see the traditional favorable development from December?

That's a good observation. Brian, do you want to --?

I'll make some comments, and Gary, feel free to chime in if you'd like. You are right, in the past several years, we have seen a fairly significant amount of favorable development, particularly in 2010 and 2011, less so actually in 2012. And partly because we actually changed some of our reserving methodology, because we saw all this favorable development in past years, so we actually changed how we set some of the reserves for each of the months, et cetera. So some of it we intentionally tried to improve our loss reserving methodology to avoid large swings of favorable development from previous years. So think (inaudible) we set separate reserves for each of the months, so -- in terms of the aging of the inventory. So some of it was process change, intentional process change.

And then secondly, for just comment on this January, which is where we saw some unfavorable development for the month, $11 million, about $4 million or so of it was in Commercial Auto, which is an area where we actually had unfavorable development throughout most of last year. So a slight continuation of that in the Commercial Auto side.

And some of it on the bodily injury side, a little bit of a higher reopen rate on bodily injury. But I wouldn't weight a lot on just the one-month development. But certainly trying to avoid the large favorable development that we had in the 2010, 2011 timeframe, we actually changed processes to try to improve that. Gary, feel free to add any other comments.

I think Brian gave a nice summary. We've changed our process in terms of how we are setting our aging factors to more closely try to align on a monthly basis. So it's a little hard to compare to prior years.

And then as Brian pointed out, within January, there was a couple areas that came in a little bit higher, not by much, Commercial Auto, $4 million, and a little more reopen activity on the Bi.

Got it. That's very helpful, actually. And my other one is a bit more whimsical, Glenn. But I was just wondering, any thoughts -- I know this isn't a near-term issue -- but I'm guessing if anyone has studied it, it would be guys -- any thoughts on what the Google car or driverless cars means long-term for the auto insurance industry? I know it is not a tomorrow thing, but if 10 years from now, it is a high penetration of total cars on the road, is that a bad thing for the auto insurance industry?

Well, we don't take that as a whimsical thing at all; we actually have a great deal of focus on those things. And sort of excuse me if I don't say everything that we think we know. It is critically important. Ultimately, the size of this industry depends on sort of really good estimates of future frequency, to some extent severity, which are harder to do. So we are very, very active in tracking almost all elements of automotive and road safety that can affect frequency.

And in fact, maybe I'll regret saying this, in our upcoming investor relations meeting, I think we will give you some insight as to how we think about those problems. We probably will not give you any conclusions relative to significant future breakthroughs, including autonomous cars and the like. We are actively involved with not only universities, but OEMs and other subject matter experts. We take that forecasting -- future forecasting very seriously and actually have some very strong positions on that.

I would tell you that while I believe the 50-year trend of reduced frequency in auto accidents, which is from a societal point of view a great thing, will continue. I don't think that takes much of a brave person to say that. The real issue is what is the slope of the line, are there any discontinuities.

The autonomous car will emerge a lot faster than it will see population on the roadways. The technology to do an autonomous car has been around for a while. We are now seeing them, we will see a lot of talk about them. The real issue is exactly how they are able to be part of the fleet of vehicles on the road in America. And that is probably not something that need keep anyone awake for quite some time.

Okay (multiple speakers).

We will take you through that a little bit --

Yes, please.

We'll take you through a little bit of our methodology and thinking on how we actually do things that are a little bit more tangible and give you a look of some things that we've done in the past that I think will give you some insight into our methodology of thinking. And you can well consider that the autonomous car, or even approaching connected vehicles, connected intersections, those sorts of things, you can assume that we are looking at them. We may not be as forthright with all of our predictions.

Okay. I guess my instinct is, if you are willing to address it, is in the sort of immediate term, when it's a small part of the population, it is probably a benefit from the better frequency. But if it ever got to a tipping point where it was a large part of the cars on the road, it is probably a negative, just because it destroys so much demand because frequency would fall off a cliff. Is that the right way to think about it generally?

I think that's a very fair scenario. And I say scenario because it takes a lot to sort of have enough of that tipping point so that the population of vehicles all behave in a controlled manner. But also understand there will be a level of threat and issue that, frankly, boggles your mind. Very few networks, very few connectivities work 100% of the time. And what does that really mean? What does it mean in terms of liability? Who has liability? What does it mean in other potential issues for insurability? There are elements of that model that might create opportunities for insurers, and certainly there are opportunities for big data sets.

So I would say to you, yes, but you might be surprised when you add the complexities of what that means, including even network reliability. It has all sorts of tangential issues.

I appreciate the answer. A very interesting topic. Thanks.

Adam Klauber, William Blair.

Thanks. Good morning. As user base technology gets rolled out to more and more insurers who use it, longer-term, does that have the potential to depress pricing in the better drivers and ultimately depress margins for that segment?

I think if we sort of start with the premise that the entire -- if one company had the entire book of business, the net premium wouldn't change, but the distribution of premium between the better drivers and the less good drivers might very well change, the whole fundamental concept of segmentation. So you could very much assume that those whose pure premium or real premium -- let's just call it the real best estimate of what they should be paying -- yes, should go down. I think we have more than enough evidence to suggest there are people on the road today that are paying a little bit because they are indistinguishable from other people like them that deserve a lower rate.

Margin go down (technical difficulty) delta would get pushed to people who are higher consumers of loss costs and [better] assign the premiums to those individuals. But unless there was some overlay factor that reduced losses in general, I wouldn't assume that the market changes a great deal at all. It's a reallocation of the cost. Take that on to a competitive marketplace, and you can apply the competitive dynamics of knowing the information before your competitors and so on and so forth, something that obviously we are very intrigued by.

Right. That would assume then the heavier users of loss cost would have to pay more for insurance down the road to compensate for the better drivers paying less for insurance. Is that right?

That's fair.

Okay, thanks. And another question, just sort of a numbers question. Looking at cash flow from operations at the parent, in the last two years, it has come down from roughly $1.1 billion to $670 million. I guess why is that and should we expect that to reverse going forward?

Brian or Jeff, you want to take that? Give us a second to catch up with your specifics.

Sure.

Are you talking about from the cash flows from the consolidated statement of cash flows or just --?

No, from the parent.

From the parent?

Yes.

The primary (technical difficulty) less than 2012. I would say that's the primary driver of it, is just sort of the underwriting profit, is the primary driver.

Okay, and (multiple speakers).

As margins change, that would be the primary influencing of it.

Okay. And then one more, just a quick detail question. When I look at the reserves, it looked like you had prior-year favorable development of $85 million for 2012 calendar year. In general, is that coming from more recent or older accident years?

I'll let Gary comment on that.

This is Gary Traicoff. That is coming from mostly older accident years. 2011, we developed unfavorably, and that was mostly on the auto BI as the severity costs increased, as we've talked about. And on the older years, we had favorable development. And we've adjusted our factors by aging to try to get a more equal distribution going forward.

Okay. That's helpful. Thank you.

John Hall, Wells Fargo.

Good morning. Glenn, you take a lot of organizational pride in your customer experience and the claims process. And I guess Storm Sandy is the largest event that we've seen since you've been distributing homeowners through the Progressive channel, as it were. Just wondering if you could offer some sort of a report card or maybe some observations that you could take away from the performance of your homeowners partners from that event.

Thank you for the first comment, because you're right, and we don't do it just as a nicety; it is really very genuine. And your question is a great one. I'm not going to be able to give you as good an answer as you might like. Feel free to ask me again on that. The reason I'm saying that is we are doing that study right now.

We do know that our NPS scores dropped as it related to our PHA experience. And I guess we shouldn't be overly shocked about that because of the nature of people sustaining a significant loss. What we need to know is was that drop somewhat consistent with overall themes? Was it specific to us?

And I mean it, please go free to ask me again. But we are doing that exact question for ourselves to assess a little scorecard on our partners and how they managed through a specific event. We have had other events, and we have had post-action reviews on those events, and our early indications would tell us, notwithstanding the NPS scores, not heading in the direction that obviously we would like or staying stable, we would add -- the earliest takeaway is that, in general, our partners performed very well.

Great. Appreciate that. I'll follow up in May. Thank you.

That was our final question. This concludes the Progressive Corporation's investor relations (technical difficulty) calling 1-888-566-0574, or can be accessed via the Investor Relations section of Progressive's website for the next year. Thank you very much for joining. You may disconnect at this time.