前進保險 (PGR) 2016 Q1 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 Quarterly Report on Form 10-Q and the letter to shareholders, which have been posted to the Company's website, and we 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; Tricia Griffith, our Personal Lines Chief Operating Officer; and Bill Cody, our Chief Investment Officer. 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 2015 Annual Report on Form 10-K 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. Nicole, we are now ready to take our first question.

(Operator Instructions). Josh Stirling, Sanford Bernstein.

Good morning. Thank you for taking my question. So, Glenn and team, I'd love to ask a big picture question on growth. You guys have had a really strong really probably six months here with growth rates rising. I'm wondering if you can help us a bit disaggregate it because there's a lot of moving pieces. And I think the big picture question I'm struggling with is how much is a cyclical story because of the environment, and how much is structural and the impact of the various initiatives.

And so when I think about it, it would be really helpful to get your color on how much you think is a function of just the competitive environment changing around other folks taking pricing, maybe on the first point, but then you obviously have a lot of retention initiatives underway is the second. And then you've got new products, whether it's Snapshot or homeowners. I imagine these are having some impact, probably more Snapshot than homeowners at this point, but would love to get it from your perspective. As we think about looking forward, how much of this is a cyclical upswing in the growth rates, and how much of this is the impact of big, powerful initiatives working their way through?

You did a nice summary of our strategy there. Yes, you are right. So I'm not sure I'm going to get as specific, but think about our October IR meeting as a better opportunity for us to address a lot of those issues. Frankly some of them are a little bit tricky to put cause and effect to, but growth is actually -- boy, sign me up for another first quarter like that one. I know we certainly had some hail at the end, but frankly we also missed some winter, so no great concerns there. And 94.6 or whatever, at the kind of growth rates we put together, nice quarter.

Why? The fact is it's certainly not luck that other competitors take rate. We try to take rate at the right time to make sure we are always positioned. I used the analogy of the wave last time. We try to ride that crest of the wave. That's sort of the underlying core of everything we do. So forget all the other initiatives. We've got to run the base business exactly the way we want to run it and keep it at a point where we will always be in a position to take advantage of market conditions when they swing our way.

We expect to always grow. That's part of our two-pronged, most important statement -- 96 -- and grow as fast as possible, but when we get that opportunity that others are perhaps scurrying where we have perhaps got ahead, we will take those opportunities and welcome them when they come.

Then the second piece is you've got to manufacture your own future growth. And the kinds of things with Snapshot, you are more than familiar with Snapshot, representing now almost close to 2 million policies being rated with Snapshot fundamentals. So you've seen from the policies that I put in my letter, that's a material part of our book and you know it's more material on the direct side than on the agency side. We've got some nice initiatives that we'll continue there with Snapshot.

We've talked last time briefly about the app, which we think can also give us an opportunity to reach out to get some other customers that are perhaps not as comfortable with a dongle inside the car. We have announced something with General Motors that will kick off within about 10 days where we'll be able to actually collect vehicle data with General Motors.

So we are starting to, as I've always said, we are starting to worry more about just the data, the algorithms and the impact on rating, less about the method of collection. So Snapshot has got some real gas in the tank and we will continue that.

Obviously, our brand strength, we continue to push on that and we've been able to do that in an environment where others have perhaps pulled back just a little bit on their advertising, so we are seeing some nice brand strength measures move in the right direction.

You flip over into the agency channel, actually I will stay with both for a while, our product advancements, we've talked primarily about 8.3, but we are always in the hopper for 8.4, 8.5, so we are continuing to develop that.

Probably the biggest move in the agency channel other than rate competitiveness is our bundling initiative that we really have done so nicely with in the direct side, lots more gas in the tank there where we use a lot of carriers. In the agency side, you know the story on ASI.

That is not a big factor in agency growth right now, so I wouldn't overplay that, but to the extent that we talk this time next year, I expect that to be a bigger factor in the growth.

So growth initiatives are all on track, different stages of maturity. I would put Snapshot higher. We've started to see agent acceptance. I talked about that last time when you asked the question. We started to see the more recent states with agent acceptance, frankly, close to double what they had been previously, so we are starting to see that move in the right direction.

So Snapshot moving along very well. Platinum in the agency channel really starting to get a grip, and we'll talk about that more in our October meeting. And as long as we are very diligent with the pricing, clearly, we see and we released numbers for -- at least you can get a flavor for what's going to happen in April. We are still a few days away, of course, for closing out April and getting you the information, 8, 10 business days. But we know that's not going to be a great month. We will take a look at our rating overall. And I would say that probably we are still in a mode where we can think in terms of 4 to 5 in terms of annualized rate increases, which is a nice place to be. We gave you some indications on trends.

So that core should keep rolling and the initiatives are doing well. I didn't answer your question specifically in terms of percentages relative to growth, and that is hard to do. We have some feel, but I'm not confident to push that out to a greater audience at this point.

That's really helpful, Glenn. Thank you. I wonder if I might ask a bit more sort of a numbers question then. I think the market has been worried about frequency for a while and I guess it's been a bit more benign.

I guess, one, would you characterize this as mostly a letup from favorable weather, which seemed to be sort of implied by your Q and maybe some other competitive commentary, or do you think frequency trends have really peaked now and the industry can now start maybe talking about the possibility of writing severity, which I would love to get your commentary on as well?

Yes, always tricky with the variation there, but I am happy to comment on that. A little piece of the story would be even different than my comments last time around. Frequency is pretty benign and certainly when we look at the BI, which I tend to worry a little bit more about, BI frequency than anything else, we are really in pretty good shape there, so we are not seeing anything dramatically go away. I would tell you wait another conference call or two to get too concerned about commentary on PIP. We've got some issues in New Jersey that we think may be driving that number a little bit differently than the true read -- I won't say true read, true long-term read.

So severity really is the play and for the most part not too much of concern there with bodily injury. I think the interesting point for severity is really the collision, and I think it's a little bit too early to suggest that we -- we clearly put in our Q that frequency was moderated by winter weather this year in the Midwest and the northern states. That's absolutely the case. We also discussed last time that, based on our Snapshot measuring, we were not seeing -- the last quarter of 2015 was overlaying the last quarter of 2014 more on miles driven, vehicles miles driven. That's changed in the first quarter. So we are actually seeing an uptick now on vehicle miles driven in the first quarter of 2016 relative to the first quarter of 2015. So that's interesting and that tracks very much with public gasoline demand as well.

So while we all speculate on the effect of increased miles, and I've always cautioned that you need to know what kinds of miles and we are seeing the longer trip miles be the primary driver of that, we may see a little bit of a tickup in frequency and we certainly need to watch that for what I will call short to mid-term pricing, but we may, we may, and please stress that, we may really be seeing a more structural long-term change in lower frequency.

As we know, cars are getting safer. We've always talked about that. That's very, very hard to measure on a month or even quarterly basis, but it would not be surprising to me to see in the future that we will see a long-term structural decline in frequency and quite possibly -- as history has shown us -- a similar offset on the severity.

So those would be my primary issues there. Right now, frankly, given that there has never been a time in history there hasn't always been some puts and takes on frequency and severity, this is a reasonably predictable and mild time. The only outlier perhaps is a little bit of PIP severity for us right now and a little bit of collision severity that we will take a look at. But that, sort of from a pricing perspective, is being offset by frequency.

Great. Thank you, Glenn. Appreciate the comments.

Brian Meredith, UBS.

Thank you. So, Glenn, I'm just curious, you've owned ASI for about a year. Obviously, some high cat loss activity in March. Sounds like in April there is also. Is it higher than you would have expected, and as a result, any thoughts about maybe changing the reinsurance program, or something you can do to maybe mitigate some of that volatility?

I will answer the first question last. No thoughts. That doesn't mean we won't ever have thoughts; we are always going to consider that, but think about the percentage of the overall book of business that we have. We currently have a small quota share percentage. We have a single attachment point of about $45 million for any single event and we have an aggregate attachment of $175 million. Those feel very much in proportion to things that we are willing to handle.

Do I like volatility? Of course not; no one does. You don't, I don't. But is this sort of in the realm of understanding of what likely could happen when we did something like this? Absolutely. It certainly also plays to the fact that we are writing an awful lot of homeowners, not necessarily with somebody's results that we have to report in ours. So the partners that we have, they are all obviously having those problems, so it's a different dynamic when we have it in the place that we absolutely need it. And as I said in my letter, it hasn't even been a momentary thought to be of any concern whatsoever, and there's no pressure from my perspective to take any less reinsurance, but, equally, no more reinsurance.

Great. Thanks. And then just quickly, on the commercial auto business, the strong growth that you guys have been seeing, some of it I know comes from some new product that you have out there. But I imagine there's also some that's going on from the dislocations that we are seeing out in the marketplace, particularly from some large commercial auto writers. How long do you think that continues for?

You are right in your summary. John, you've had a look at this. Do you have any -- I don't have a really strong view because I can't know what the other competitors are doing, but do we have any other insight?

Sure. Yes, we have seen a lot of dislocation and we think we were well ahead of the marketplace even as much as two to three years ago and seeing some trends that we took some material rate to address and also added some significant underwriting efforts to a lot of the lines we write there to ensure we were writing business that was going to be profitable, so we are really well-positioned now.

And, yes, we are seeing a big increase in quote volume. So if you think about new business incoming as either a function of an increase in conversion or quotes and in commercialized, it's predominantly in quotes, which generally means competitors are raising rates, or just not choosing to quote the business. As far as where that goes, we can't know with certainty for sure. We are confident in our rate level.

In some of the areas, we might take rates up a bit, but obviously you see combined ratios there that are very good and we expect will continue to be very good, so we are pretty bullish on growth moving forward for commercial.

Great. Thank you.

(Operator Instructions). Gary Ransom, Dowling & Partners.

Good morning. I noticed you gave some attention to your new relationship with Uber in Texas. And it seems like there's a big increase in the potential for the data that you could collect in your understanding of driving behavior. It's not just individual drivers like you are doing with Snapshot, but it's tracking a network. Every one of these drivers as a GPS and you will have your own app that you can combine with it. I'm just wondering if you could give us a little more color of how much this might change your view or how you can look at driving patterns, driving behavior out there.

Gary, (inaudible) to right now, you know I'm not dodging that question, I'm genuinely not, but the fact is your thesis is so on point, just understand that's sort of what we are all about and when you get a situation like Uber where you can truly collect the data, there's a wealth of data whether we incorporate Snapshot or whether we use the data that's otherwise available to us, this is truly an exciting situation because it just plays entirely into everything we feel we are good at and hopefully we are playing into one of Josh's comments earlier, just one more thing to have another arrow in the quiver for places to see growth coming later.

But there's not a lot to report right now and I'm simply not dodging your question, but the idea that this is just an insurance program, it's much more than that and that's why I used the analogy of the square peg, round hole. And I think we have really developed a program that makes a lot of sense, and hopefully the way this is really successful is that we know what that vehicle is doing, where it's going. So the route intensity, braking intensity, all sorts of things that we can know, not only at the fleet level, but we can know it at the driver level as well.

So frankly this is one that, if I was in person, you would see me smiling because I think this is one of those opportunities that comes along relatively infrequently, and this just seemed like it had our name written all over it.

My imagination may not be enough to see other opportunities of a similar nature, but are there other opportunities out there where you can gather more informative data like that?

Yes, I think you've got to accept strategically that that is the mindset you've got to have. You can take it all the way from the private passenger automobile, which we are doing slowly with Snapshot, but you will see taxicabs being more involved in this type of thing. There was a release today in the Wall Street Journal today about GM working with Lyft. We have a relationship with Lyft. There's a lot more vehicle-sharing. It's a minute part of the economy right now, but certainly one that if you were a betting person you'd probably bet on seeing that being a lot bigger later.

We want to make sure that we are designing contemporary style insurance programs for those kinds of options, and frankly, I think we are extremely well-suited for it. We know others will have some interest, but we are doing an awful lot of groundwork right now and we are very happy with that. So I think this is going to be something we are going to be talking about a lot more in the years to come.

Thank you, Glenn, for that color.

(Operator Instructions). Meyer Shields, KBW.

Thanks. Good morning. Glenn, you mentioned I think in your response to Josh that you expect increases in severity to offset the frequency decreases that you can anticipate from improving cars and that kind of surprised me because I guess I wouldn't have necessarily seen the historical connection persist, and I was hoping you could talk a little bit more about that.

Well, we have actually shown in at least one IR meeting, I'm not quite sure whether it was two or three years ago, what I would call, if not a perfect monotonic function over some reasonable period of time, a declining frequency. The fact is we've made our cars safer in this country for the last 30 years and there's nothing on the immediate horizon that suggests that we are not going to make them continuously safer, some of the things even the federal mandate for emergency braking. Those sorts of things, it's impossible for me to think there isn't a real macro driver to yet make cars safer. That's my premise for declining frequency.

What's interesting is that, over that time period, the market for private passenger auto has grown, and that is largely a severity offset to the -- actually it's even more than an offset, it's a gain -- over the frequency decline. And those are interesting and accurate and observable facts when you start to decompose it to try to say exactly coverage by coverage what's driving it. That gets a little trickier. But I would -- I'm working -- at least my mindset, is that a long-term strategic view is that we will see fewer accidents and more expensive accidents.

Okay. That's fair enough. And then second, sort of a detail question. The jump in the commercial auto expense ratio in March, is it fair to tie that to the new Uber relationship?

No, I wouldn't tie that. That might be a penny or two there for sure, but actually there's probably two bigger drivers of that. We are working to put a great new system in for commercial, so there is some expense associated with that. But mostly it's driven by advertising costs as we continue to grow our penetration or our attraction for direct operations so that landscapers and small business people who have needs that are not that dissimilar to private passenger auto and choose to operate or interact with us on a direct basis, we are building that out and that definitely seems like the demand is there, so we are doing more advertising to support that. That'll be the biggest driver of the expense difference.

Okay. And that sounds like it will continue based on your comments on the environment?

Yes, it'll be a little spiky from time to time based on seasonality, but, yes, expect that, but that doesn't mean there won't be some equal and opposite offsets as we build that retention book in direct. So a lot of it is front-ended.

Okay, perfect. Thanks so much.

(Operator Instructions). Ian Gutterman, Balyasny.

Thank you. Glenn, I just want to follow up real quick on the frequency comments in the Q sounding a little bit better and then your comments earlier about miles driven being up, which we can all see. Usually those two seem to correlate a little bit more. Seems maybe they went in opposite directions this quarter. Any thoughts on that?

Yes, I'd be careful to make a quarter-to-quarter comparison there. Let's take a look together at next quarter or even next six months to see how the frequency is ultimately being reflected through. It should be a pretty close comparison, but I'm not always convinced that it gets 1 for 1, so I'd wait just a quarter and see.

Got it. Okay. Great. Then a couple things on the homeowners. One is just can you just tell us a little bit about how your reinsurance is structured? And I guess what I'm getting at is I'm guessing, because a lot of others have done this, you probably have some aggregated cover and just given the March events and the April events, how do you stand with that aggregated cover? What's the risk of essentially going through that if we have more events later in the year, especially a hurricane? Do you need to buy live cover for the rest of the year?

And then just big picture on your home, just from what I can see in marketshare obviously, your two biggest states are Florida and Texas, which are heavy cat states. As you roll out the Platinum, is it a specific goal to grow away from cat areas, or is this sort of take it where you can get it if it's a good customer and you will sort out the cat mix later?

Well, we try not to do things later; we try to think about them ahead of time. But the fact is Florida and Texas, that's where people live, so we are going to be active there and we are comfortable with it. And the reinsurance that ASI has put together right from the beginning is obviously in a position to respect the fact that they have coastal exposure.

So why don't you go through, John, the three layers? We are a long way from sort of overly any concerns that Ian has there, but why don't we at least talk about the $45 million single event and the $175 million bond. And Trevor is right here, so if we need additional help, we can get that.

So obviously, we gave you some insights into ASI's reinsurance program in the Q. Glenn really just covered them. The first layer of $45 million and the tower on the first event, we won't start quoting PMLs, but trust me that it is much higher than I would expect the average industry buys to. So going through the top, so to speak, is a highly unlikely event.

And, yes, we then have a catastrophe bond facility that is at an aggregate of $175 million, and that is for storms throughout a calendar period. We are covered on second events, similarly, not quite as high a tower, but we also pre-purchase reinstatement on many of the layers. So not completely up the tower, but unlike many other players, if we ever had to reinstate, those premiums are prepaid for much of the reinsurance.

So ASI has an extremely robust reinsurance program. All of the players in that program are virtually all A-rated or better, so we are very comfortable with the program. And that $45 million, they had targeted previously that retention to ensure that at any one event less than 10% of their surplus was at risk, and obviously less than 10% of their surplus is a pretty small percentage of the combined entities, so a pretty robust reinsurance program. Trevor, I don't know if you want to add anything to that or --?

No, I think you nailed it pretty well.

That's great. Very helpful. If I could just clarify, Glenn, on the Platinum part again, just is there any restrictions at all? Do you say we'd rather not grow Houston if we are going to grow Texas, or we'd rather not grow Miami if we are going to grow Florida and maybe just market it more aggressively in the Midwest, or the Northeast or other places where you have a little bit less cat risk?

ASI has historically controlled the risk to a significant degree in Florida, only taking the amount of business they felt comfortable with. We are very conscious of the concentration layers, but as we roll out the rest of the country, obviously that will become less dependent on any one state. So there's a yes and a no answer in there. Yes because of where we are today, but ultimately we want to be a meaningful bundled option for consumers and Florida is a big state and we will, as you just heard, make sure that we have all the right reinsurance. We are not going to go crazy. We know our objective here is to try to get customers and have them with us for a long period of time. And if we are more than comfortable letting someone take the layers above that, and hopefully they will make profits on that over a reasonable period of time because we will achieve our objectives as well.

But don't assume that somehow we are going to only be marketing this in the interior states or something like that. We are going to be as aggressive in states as we can be. But hopefully you would agree; we will be very thoughtful about that and have the right level of risk-sharing. That doesn't mean, and this is probably interesting that we have sort of got the layers that we have, we've got the results that we have, this may be sort of the pain point right now in terms of absorbing the results and the real question is how do we feel about that volatility.

I think we are getting -- I wish it wasn't the case per se -- but I think we are getting a sense of exactly what that volatility can mean to us. And my only encouragement -- it's just my encouragement -- is given that we are a monthly reporter, you are going to see that volatility just be a little bit more dramatic than you would over longer periods of time. There is not an ounce of us rethinking our strategy or doing anything differently at this point. I shouldn't make it flippant in that case. It doesn't mean we haven't bought a great deal about it; we are just coming back and confirming the positions that we have.

Perfect. I want to make sure --.

I might add that ASI did enter California. In January, I believe, we started writing business -- February -- and we expect to be in New York mid-year, May. So naturally as they expand across the country, you'll see less concentration in those coastal states. But, to date, yes, they've been pretty highly concentrated in the Southeast. But I think naturally as they enter more states where there's a large populace as well, you'll see that concentration decrease.

Understood. Thanks for the help.

Mark Dwelle, RBC Capital Markets.

Good morning. Just wanted to get a little bit more information on the cat losses you've already reported for the second quarter. Were those primarily concentrated in the Texas market, or was it a little bit more geographically dispersed than that?

Jump in guys here. I think Texas is a big driver. Even if you watch television, you'll see that Oklahoma has got some things. A little in Kansas, but the real driver is Texas.

Was that more skewed towards the flood losses in Houston or the hail losses in San Antonio?

Well, let's break our two lines apart. Frankly, for vehicle -- well, it's the same real issue. Hail and driving weather really have been the cause for both auto and home. Home is not really affected by the flood conditions in Houston. Some of our vehicles are, although that actually was a relatively manageable number. My memory tells me about 450 losses directly related to the flood, so that's not a big driver of our results this month.

So when you see that 135, the 85/50 split, think about the really nasty peril of hail. When hail gets to be the size that it is, it is warfare out there for roofs and for sheet metal and there's a lot of damage. That's just plain and simple. Hail is one of the most nasty perils that we come into. So comp is going to be dramatically affected by that and, as I said, we will take a couple of opportunities, even though the numbers will be what they are. The first thing that we will do is recognize that these are first-party claims.

So we are going to pay the loss, so we've got to turn that into a marketing expense. This is the time that we get with our customers and make their lives better after an upset, whether it's their roof or whether it's their windshield or whatever. So the first thing, the attitude we are taking right now is turn that into a marketing expense. We know how important retention is. We know the relationship when you give great service and claims in the future retention, so that's the first thing.

Second is take a look at our pricing, make sure that we are where we want to be. Certainly one spike of a month, we've seen that on numerous occasions. We are not at all concerned that we are off track for our commitment to shareholders, our 96 or better, but we will reevaluate our cat loss or our cat loads. And I would to you if I was looking at the overall market, I see rate change on an overall basis, or at least driven by big competitors.

The last data point we have, February, around about a 4.5, which is about 110 basis points stronger than earlier in the year. My suspicion, based on knowing what's been filed, we will get more clarity on that from competitors for April and May, but we know there's been a lot of rate changes filed, so you could see that go a little higher.

And I would say if I was giving you direction last call, I probably would've said think in the 4 range, so think about a point per quarter if you average it out. I'd say put another 25 to 40 basis points on that per quarter as we look at the rest of the year. And we might be a little higher when you factor in cat loss, cat loads, but for the most part this is all quite manageable for us.

That's really helpful. Thank you. One last nerd question, when events like these happen, do those tally into frequency and severity statistics, or is that a different calculation for cat events like these?

They are mostly comp at least on the vehicle side. So typically when we give you numbers for frequency and severity, we don't put comp in there because it's -- that would just be a random walk in terms of your graph. So most of the frequency and severity numbers we give you are the more sustainable coverages like BI, UMBI, PD, but certainly as we look at it, yes, it factors in.

All right. Got it. Thank you very much.

(Operator Instructions). Bob Glasspiegel, Janney Montgomery Scott.

Good morning, Progressive. Glenn, I'm going to push back a little bit on your slight warning about miles driven potentially having an impact on frequency. Gas prices are up 30% from roughly the average level of where they were in the quarter. We are going to soon be lapping year-over-year impact of gas prices. If lower gas prices hasn't really driven a big impact on frequency to your book to date, why just in the future should we be incrementally more nervous about it?

Well, I'm probably going to [rate] myself here a little bit. The last quarter, we had our -- 2015 fourth quarter wasn't that much different than 2014 fourth quarter. As we start this first quarter of 2016, we are starting to see that be considerably above. That tracks with gasoline demand. That all makes reasonable sense.

Now the question is how high will it go. Well, people aren't going to drive infinite miles just because gas is cheaper. They are not going to drive to work more often, so this is more recreational and discretionary type trips. And frankly, I think everybody can have a theory on that. I would just tell you that what has changed is, in the first quarter, we are seeing miles driven as our Snapshot population, which is large enough to be credible -- I'd make the claim that is different than the claim I made for the fourth quarter where we hadn't seen it be dramatic; in fact, it had flattened off. It is now showing increase and we are going into what is really the high vehicle mile traveled part of the year. So it'll be interesting to see whether or not the discretionary usage continues to be higher or not.

In this one, Bob, I think all I can do is report the news. We have incredible interest in this sort of thing because it makes a big difference. But whether I could say that we are seeing a dramatic change in frequency simply because of that; no, we take it as it comes and the indications that because those miles are more discretionary and longer distance trips that the effect on frequency is not quite as linear as you might expect.

Totally with you. I guess I was asking more a futuristic question, that the data that you were looking at is when gas prices were down a whole lot year-over-year, and we've had a 30% spike in gas prices and I'm just saying if you were looking at real-time data the last few weeks, I would think that you are not going to see the same year-over-year mileage increases that we saw in the first quarter where gas prices were down a whole lot versus a year ago.

I wouldn't bet against you on that one. I don't have the April numbers. Actually, as soon as I get off this call, I'm actually probably going to get more of the April numbers. That will be very interesting, yes.

And I would add, while we can't model it perfectly, we think gas prices -- there's some step functions in there. Certainly back when we hit $4, that was a ceiling that people retracted a lot, but we think in some reasonable range, sort of the marginal change in driving is not that linear with gas prices.

Certainly economic. And trucks on the road -- this is a very complicated trucks versus trains can obviously impact it a lot.

Airline travel too for discretionary travel, yes. Unfortunately, we report the news more here, but we have our own theories, but I don't dismiss or suggest that yours is not an interesting one as well. April will tell a story because we are really going into April and May and June, which are the high travel vehicle miles traveled months.

Right. It looks like you lengthened your maturities a little bit in the first quarter, took your short term down. Anything material behind that and anything on the investment side that we should be aware of that you are doing?

Bill, do you want to jump in on that?

Sure, Bob. Yes, actually, we shortened the duration of the portfolio a little bit in the first quarter as rates fell. The short-term change or the big increase in short term in the first quarter is really not so much a portfolio strategy change as it is a definitional change.

What I tried to get to in the Q is we had about $1.1 billion of short-coupon treasuries that counted in our treasury portfolio that matured in the first quarter, and we rolled those into treasury bills, which are counted in the short-term portfolio. So no real big change there other than a 10th of a year decline in the duration of the portfolio.

The other changes that we had was we increased our corporates a little bit in the first quarter as we saw some more compelling opportunities when spreads were wider. And then we've made some little repositioning around the ARX portfolio where we took control of the management of that portfolio on January 1. So there have been some turnover in regard to that. But other than that, no material changes in the portfolio and certainly no changes in our overall strategy there.

Thank you.

It would appear that that was our last question, so that concludes our call today. Nicole, I will turn it back over to you for the closing script.

Thank you. That concludes the Progressive Corporation's investor relations conference call. An instant replay of the call will be available through Friday, May 20 by calling 1-800-253-1052, or can be accessed via the investor relations section of Progressive's website for the next year.