前進保險 (PGR) 2015 Q4 法說會逐字稿

Welcome to The Progressive Corporation's Investor Relations conference call. This conference is also available via the audio webcast. (Operator Instructions). 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 the letter to shareholders, which has 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 our 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 the 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'll find discussions of the risk factors affecting our businesses, Safe Harbor statements related 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.

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

(Operator Instructions). 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. (Operator Instructions).

Bob Glasspiegel.

I was very encouraged by the last quarterly policy expectancy lengthening that you highlighted in the 10-K. How excited are you by that trend? Do you think that this is a breakout that suggests that Snapshot and some other measures you've taken are working?

How am I excited about -- I been waiting for it for a long time, so I'm actually pretty excited. And I think the 5%, thereabouts, that we are seeing in the trailing three is really quite indicative of what we hope will flow through. Lots of reasons for that, always hard to pin them down; but general excitement, absolutely yes.

If you take a look at some of our published materials, you'll see that the average rate that consumers have seen over the last year, regardless of channel, agency and direct, is about a 4%, which I would tell you to think more in those terms than specific rate changes. So, 4%, which means a good number of our consumers are still seeing rate changes that are on the high side of that.

So as we mature into the rate revisions that we are now becoming even more comfortable with, and rate levels that we're becoming more comfortable with, I would look forward to seeing, while continued positives, maybe not quite in the 4% range, but a closer distribution or a tighter distribution around the average. So we will potentially benefit even more so on retention, not losing those higher end or the right-handed distribution around the average of 4% last year.

So really nice movement, getting our new product into the marketplace, getting a rate level that we feel very comfortable with, maintaining that. You might remember, we said at our IR meeting we had a strong preference for three 1% changes than one 3% change.

All of that really is a big part of how we're trying to make sure that retention is front-and-center in all of our thinking. So, nice to see it coming through. Frankly, I would tell you, I'll be disappointed if that didn't continue.

Okay. Switching gears. Bill, the first quarter we've seen a stressful investment environment. Are you maybe going to update us on how the fixed income has performed in the stress lines in -- are you taking this as an opportunity to increase risk? Or are you de-risking on the portfolio?

I'll let Bill jump in there.

Sure. We are taking this as an opportunity to increase risk modestly. We've increased a bit in the corporate exposure, a little bit in CMBS as well. Some areas we have decreased risk, and that would be primarily munis, because those have traded really tightly and very opposite to the rest of the fixed income markets. And also we have reduced some of our high-yield exposure, because the high yield that we own has performed really, really well, to the point where it's trading at spreads similar to BBB paper. So, for us, we are in a good position to take some risk, and we're finding some more compelling opportunities to do that.

Thank you.

Performance-wise, we released our January numbers, which -- or our total return, we will release February, in a couple weeks.

(Operator Instructions). Josh Stirling, Sanford Bernstein.

I appreciate you taking the call. I'd love to ask you a question about the status of Snapshot. Over the past six months or so, you guys have announced you've been rolling out a new smartphone app. And I think at the last annual meeting, you walked us through that you were iterating the pricing design to add discounts up front.

And I was curious, just one, a bit of an update on how many states you have rolled these things out with. As well as, more importantly, what's been the consumer response, both from the app, which feels like a different consumer interface? And is a question of adoption and comfort and more open-ended questions about how consumers like the app and whether perhaps something will sell.

And then separately from the discounts, what kind of lift have you seen from this? Is there an impact on conversion rates and maybe on new sales? Thank you.

Yes. We haven't done a little Snapshot update for a while, so let me see if I can hit all of those points and give you some flavor of what's going on. Starting into this year, we are still seeing our policies initiated with a Snapshot influence -- discount, whatever it might be -- to be faster than the rates of growth that you are seeing as we report on aggregate. So it is still our fastest-growing sector. That remains true for a long time; but certainly, even into January, that was true.

Starting to get a feel for what sort of mix of business is attainable, so I'll just give you a quick look here, not to confuse you with any numbers. But in our direct book of business, where we have been able to offer Snapshot -- and realistically there's only two states now, California and North Carolina, where we are not offering it. But over a period of time, where the cohort or the Snapshot has been available, that cohort of business is now approximating about 40% of the direct business for places where it was available for some period of time.

So, think in terms of -- can the little train really take momentum? We are looking to sort of 40% to 50% in the direct channel. I think I've told you before, the acceptance in the direct channel far exceeds the agency channel. And with the new change that we made -- I was very clear that we were hoping to see agency uptake, or on the -- or the quotes include Snapshot with that discount (technical difficulty) what I'd call overly dramatic movement, but significant, on the small base.

Actually, we have some good news to report there. We actually introduced it into two what I'll call higher average premium states -- Florida, Louisiana. And we're seeing the agent acceptance rates there, or the agent penetration into new app or new quoting, actually go up significantly. So probably approximating somewhere in the league of one-half to two-thirds the rate on our direct channel. So, starting to make some traction there, and we're going to take a closer look.

I think we did a better job of rolling that out to agents in those states. So we'll take a close look at what we did well there to make sure that we reinforce that in other places in the country. So net-net, pretty good-sized penetration in our direct book. And I don't want to project what the threshold might be as to overall penetration, but to think in terms of more than half the book is not unreasonable based on what we've experienced to date. And now we're starting to see opportunities where the agency business perhaps is, yes, slower, but certainly starting to close some part of the gap. So, good stuff there.

With regard to the app, we've discussed that. I want to make it very clear that we've had apps that have approximated, in one way, shape, or form, the activity for some time. That's not been an issue. The issue is really making sure we had an app that we felt had the data integrity that really matched with the in-car dongle. And we've got to the point where we actually have that application, and very comfortable with it. And there are all sorts of stories that -- some of which I'm not as close to as others -- but you need to be able to determine the difference between a train ride and a plane ride and a car ride.

You need to know whether the person got in the left door or the right door. Are they driving, are they a passenger? Although sorts of things are actually very important to us. So it's easy enough to create an app that does part of the job. It's very hard to create an app that does the -- or gives us the level of integrity that we'd be comfortable swapping out for the information we get for the dongle.

We think we are there. Unfortunately, it's probably going to take us a little longer to get that into marketplace than we might like. There's a fair amount of work to do on the IT side. But certainly in 2016, we expect that to roll out. But it may be -- for all intents and purposes, in large numbers, later in the year.

I think we've probably mentioned we have a relationship with GM and OnStar that's been a little bit delayed, but we expect to be marketing that more actively in about the mid-year time frame.

So Snapshot, for lots of reasons, continues to be something very important to us, not only as an indicator of the sorts of things that we think we we'll do more of in the future; but it's really taking hold, and we all know the power of Snapshot relative to a rating variable. So, ultimately, when we start talking about penetration numbers that are really in the range of that 40%, perhaps even 50%, it starts to make a difference in terms of segmentation.

So, we couldn't be more happy with what we see. I just don't know what the high-water mark will be, but we're going to find out, and I hope it's really high.

Thanks, Glenn. That's really helpful. I wonder if you could just touch briefly on your other big product development initiative, rolling out a combined homeowners product in the agency channel. Give us a sense of -- I think you are live in at least in a couple of states. It would be useful to have a little bit of color for how agents are responding, what kind of traction you're getting. And then maybe if you could go into -- walk us through what your goals are for the program as you look out over the next year.

Sure. And maybe I'll double-team here with Tricia on a couple of points. Platinum is what we're talking about. And this is, in short -- I don't want to repeat things I've already written down -- but, in short, is the way that our ASI property business in Progressive come to the market looking as if they are simply one. So, let's not worry about the brand names, but the agents are seeing that as Platinum. We are in 12 states. We expect to be in an additional 22 this year, so that will give us some really nice penetration.

Our goal? Our goal is simple. Our goal is to be a valid and almost significant choice inside the agency channel for those customers who want to bundle their auto and home together. We're comfortable if they start either way, but we really like the bundles. ASI is seeing about 50% now of their new quote volume being bundles. So we really -- we're not going to curtail, in any way, shape, or form, monoline of either. But our real interest is in the bundle. We know what the effect of long-term retention there can be. So, those things are great.

Our interest is making sure, on the agency channel, we offer a really viable option and a, as I put it, must-have solution for agents. We know we've got a great auto product. I don't think there's any dispute on that. With ASI, we have a great homeowners product. We're going to work together to put that into a great package. And there is still work to be done, so we don't think we're across the finish line. We probably never will get across the finish line. We can always make it better.

And on the direct side, well, ASI is not as big a player on the direct side. That's obviously a future option that we have for going forward.

Tricia, maybe you could talk a little bit about the salesforce, and how we've oriented the salesforce to make sure that we get the message out as well as possible.

Yes. Josh, well, I think to what Glenn said, before I say that -- we really, first of all, the second half of last year, getting the product out, the [AT] product, was really important to get that referred business, and then continue with stable rates. And then we had to have both an agent and a customer value proposition. So we know we needed to add more commission for these preferred customers, match policy terms. And then as we started to talk with our ASI partners, we realized we really needed to combine salesforces and have very specific sales reps of that work with our Platinum agents.

So we've just done that recently. We're very excited about that. And we continue to -- like, we have 12 states out -- we have continue to understand where we are competitive; when we're not, what we need to do. And I think agents understand that. We really want to make sure that we reward our agents not just with more commission, but what preference, so when they choose us as number one or two. In fact, just this week, John Sauerland and I were able to make some calls to agents that have reached some certain levels. We have hurdles to get over, determining how many bundles that you have in your agency.

So, while the data set was 12 states, I would consider 2016 really the year of execution, understanding more and more about, again, how to get that low pure premium business, and how to integrate the right salesforce with these very specific agents.

Great. Well, we'll stay tuned. Thanks so much.

(Operator Instructions). Ian Gutterman, Balyasny.

First, just a quick numbers question: when I was looking at your [stat] surplus for the end of the year, and your leverage -- [your] normally [average] closer to 3 times; this year, you are at about 2.7, which I'm assuming is the homeowners change, obviously. Just when I backed that, if I took your normal auto premiums, divided by the 2.9 to 3 times, and solve for the rest, home was coming in at about 1 times levered. Is that a good proxy?

John, jump in here. You're right; the 2.7, a little bit influenced by the home. We would probably prefer to run the home business at more like 1.5. And I expect, over time, as these things sometimes you get a little bit off; but think in terms of we're going to try to maximize on auto at about 3 to 1 if we possibly can, where we can -- never works out perfectly -- and closer to a 1.5 on the homeowners business.

Okay, that's what I was wondering. Okay, great. And then on the reserves I guess just first quickly on the January adverse development. Normally -- so I guess I'm going back in time a little bit. There used to be big [favorable] release in January due to the catch-up from the prior -- or December. And I recall you changed that process about three or four years ago, to smooth that out. And this January, we saw the lumpiness, just in the other direction. Is it still the same process? Or do we make a change in process back to the old way?

We've got Gary here. I think the IBNR with the process change -- he's coming up to the table, so I'll let him take it away. But the unfavorable that you saw in December, I would -- I don't want to be cavalier about it, but I would tell you, don't be too concerned about that. If that had been February over January, then it wouldn't be prior-year. It just happens that is the calendar close, and therefore you see prior-year development come through.

As we looked at our January development for 2016, there was really nothing that was surprising over the prior year, so we're pretty comfortable with that.

Gary, why don't you put in some more details there?

Sure. This is Gary Traicoff, Chief Actuary. To your point, about 3 to 4 years ago we did make a process change. And then that last few years, it looked fairly stable. What happened this past January was really due to more late reported claims that happened in December; they got reported in January. And if you think back to the end of December, we did have some storms come through, particularly in that last week or so. So in January, we just recognized more late reported losses, which was really the driver of the development in January.

Got it. And Gary, while I have you, if I can clarify one other thing. I was looking in the K, on the reserves it said our net reserve balance assumes that's varied for X, and your 2015 over 2014 will be 4.2% higher for personal auto and 9.2% higher for commercial auto. And I was trying to contrast to earlier in the K, when you talk about calendar year severity, when you give your frequency and severity stats. And that calendar year severity was 3% for PD, 4% collision, minus 2 BI. If I weight those, I get a very low single-digit versus the 4.2% on the action year.

Is the difference between that low single-digit and the 4.2% reserve releases? Or is there some other way to interpret that?

No, that's a great question. Related to that, when I think of calendar period severity, which is what the low numbers you were talking about, our calendar period severity that came through in the year was fairly low; particularly in BI, it was actually negative. And that was really the driver of why our reserves developed so favorably in 2015. When we think ahead, and going forward, we don't anticipate long-term severity will continue to be negative like that.

So if we look at what we actually have booked in our reserving models, we are anticipating roughly a 4% severity going forward on liability. And that will include BI, PIP, PD, et cetera. So if you do all those cuts by state and by line, most of them are in the 2% to 6% range, but overall it rolls up to about 4%. When I look at commercial lines, it is about a 9%. Some of that, though, relates to mix changes in the business marketing targets we have.

So if you normalize for that, think of it probably closer to a 6% range or so, which was actually fairly consistent with our commercial line severity in 2015.

Got it. And do you have something similar for that, what the accident-year frequency was in 2015?

We do not report anything in the per-accident-year frequency. Typically, though, calendar year and accident year frequencies are fairly well in line. You're not going to have as much of the disconnect as you might have on the severity side, particularly with the BI and the long tail coverage.

Great. Thank you so much.

(Operator Instructions). Brian Meredith, UBS.

Back on the Platinum product, a couple quick questions here. First, how are PLEs looking so far with respect to the customers you are getting? And then on that also, I'm just wondering is there an opportunity here, given the consolidation we've seen in the mass affluent market, that you could potentially go after that right now?

Brian, could you just give me the second part of the question again, the consolidation piece?

Sorry, yes. What we've seen as far as the consolidation in the high-net-worth or mass affluent market from a homeowners perspective -- is there an opportunity for you guys to really go after that? Because I know agents are asking for new carriers there.

Great. Let me -- yes, let me take the first. I think we've shown you at some prior IR meetings, and we're more than willing to show you as we get more and more information, that the bundled customer really does give us that notion of a meaningful -- and I'm going to stress that -- a meaningful extension on our life expectancy. In my shareholders' letter, I started to use some language that I think will become more common inside Progressive. Moving away from policy life expectancy to customer life expectancy -- I think you get that.

It's really we are looking to at least a piece of our business -- and I stress that; we will still always do the pieces that we do well -- a piece of our business being customers that actually stay with us in a reasonably long period of time -- or actually, not a reasonably long period, a very long period of time -- and have multiple products.

So a way to think about that -- so yes, we are seeing -- yes, that was our hypothesis. Yes, we are seeing it. And the data is solid and meaningful. I think, quite frankly, we might even be understating the customer life expectancy yet, because we haven't seen that kind of customer for a long period of time. Something just internally, and this is just to give you a little bit of a mindset of how we think, we are starting to -- and this is, I would say, something that 30 years ago was certainly was not on our radar screen -- how many -- what percentage of our customers have been with us for more than 10 years? And now we have a meaningful double-digit -- let's say that around 15% of our customers that have been with us for more than 10 years.

That wouldn't be a number that would impress USAA or State Farm, I'm sure. But relative to the repositioning of Progressive, that's a number that's very important. And the type of customers that we are seeing go into that state of decade and decade-plus customers, are the bundled customers. So, nice stuff there.

With regard to your question on the high net worth, I'm going to be very frank. We've got a big agenda in front of us, and we're going to probably play in something below the high-net-worth customer, just put it out on the table exactly that way. That doesn't mean that we won't make some phone calls and have other people's product to sell to those high-net-worth customers, at least on the home side. But that's probably not the focus that we're going to add to our property business here any time in the immediate future, given the very large market opportunity there is for us that we feel very comfortable realizing.

Great, thanks. And then just one other one. Any changes that we could think about with respect to your reinsurance program?

No. We will go through the reinsurance renewal. Trevor is actually here. But in the June-July type time frame and -- or John, correct me if I'm wrong there -- but, no, I expect that we will take reinsurance with our property business very, very similar to the way that we have taken it in the past, at least for the excess layers. We may, and likely will, reassess some small part of a quota share that really isn't necessary, from our perspective, going forward.

But in terms of the layers, we are not looking to take higher-end risk any greater than we have taken in the past, or ASI has taken in the past. And that was never the objective. The objective is to get more customers in the door. We will take a reasonable amount of risk, but we feel very good about the amount of risk we're taking and don't intend to extend that.

Great. Thanks for the answers.

(Operator Instructions). Meyer Shields, KBW.

I have a quick question, just to understand your reporting. And let me use one example. So you have indicated there are direct renewal application changes that were 5%, I think, for the year. In earlier quarters it was 7%, 7%, 8%. So should we look at that 5% as just a modest dip from the 8%? Or is there a bigger dip in the fourth quarter that gets the full year to 5%?

You want to take that, John?

Sure, I can take that. Recognize we do have a calendar year that in some years has more weeks than in others. So in last December of 2014, you might recall we had an extra week. We [referred] to that in a December release for you. So, in that release I think we also normalized the premium for the month for you, to give better perspective there. I would not interpret that at all to be a dip in renewals. The PLE is what you should be looking at to understand trends in terms of customer retention. And as we discussed earlier, we are seeing good changes there. I think you are just seeing a reporting anomaly, nothing material.

Okay. No, that helps a lot. Thanks for clarifying. And second, if we look at the property side, so if I'm doing the math right, you acquired $220 million of reserves, and released a little bit more than $30 million. Is that 15% a good representation of ASI's reserving philosophy?

No, I'm going to answer that a little bit more generically. One of the things that I am delighted about in the first nine months of being together is that we feel, and I think ASI feels, that we have some skills to offer, and clearly they have product and skills to offer as well. But one of the places that we were able to actually dig in together is on the reserving side. And we now have an actuary that has been on our actuarial staff working directly for Trevor, the CFO of ASI; but also having very direct line or dotted line, whatever we want to call it, as an integral part of our reserving group.

And during the course of that nine months, we applied some different techniques. We looked at data a little bit more specifically to ASI versus industry data. That had been done before. There was no right or wrong. There was just -- we're digging in in a way that perhaps is a little bit more intense, a little more reflective of the way Progressive has done reserving. In this particular case, it turned out we had some releases. Could have been the other way; we would have adjusted, either way.

But I would tell you that the reflection of the releases this time is a little bit more from a highly intense review, overlaying some models, overlaying some techniques, looking at data, looking at it more frequently. And what I expect, going forward -- obviously I cannot guarantee anything -- but what I expect, going forward, is we will be much more on a glide path of the kind of approaches that we take traditionally with our auto business. And I suspect the volatility will reduce.

Okay. Thank you very much.

So probably one-time volatility that is reflective of that process; but, frankly, it's a great opportunity for us to really work together in taking a look. I have to say, it's always nicer that it's a release, not an add. But, frankly, I feel much better that we've had the deep study. And everybody's very comfortable.

Okay, that's very helpful. Thank you.

(Operator Instructions). Jay Gelb, Barclays.

Now that you've gone through the year-end 2015 process, I just wanted to get your updated views on why you feel Progressive is not being impacted nearly as much on the personal auto side as a result of increased miles driven translating into worse claims frequency trends like we're seeing for essentially the rest of the industry.

Jay, I might challenge the rest-of-the-industry piece there a little bit. While notwithstanding that frequency was up, we are reporting across the board 2%, 2.5%. I don't think that is dramatically different than the PCI reporting part of the industry. Fully note that there were a couple of very large and very notable -- and competitors, who know perfectly well how to calculate frequency -- they reported higher numbers. And that's just a fact.

I can't necessarily know, other than at least in one case, probably a mix adjustment that would really be needed to be made to normalize those numbers. But I'm not sure that we're the anomaly on that, relative to the industry. And with regard to -- but to be fair, any time -- and I mean this sincerely -- any time we see anybody else seeing something that we're not seeing, trust me, the radar just goes off like crazy, and we make sure that we're not the ones waiting too long to see that kind of effect.

But as it has played through from an order perspective, we really didn't see a great deal in frequency. We didn't see any major trends. We [started] up a little bit in the third quarter. But for the most part, that 2%, 2.5%, not bad. Now, remember, systemically, long-term, we think maybe frequency is headed for a reduction, just for lots of reasons with vehicle technology.

Miles driven -- no question that that would be the usual suspect for increase in frequency. And to the extent that we didn't see it as much as some others, I would also tell you that our most recent reporting on the mileage increases, related clearly to gas prices, we're starting the second half of 2015 -- or we did see the second half of 2015 -- be a lot less dramatic of an increase over the prior year than the first half of 2015.

And our most recent data points might suggest that we have hit the threshold of miles to be driven. I think we'd be better off waiting till the spring and summer to see if we get another spring and summer effect. But for the most part, the more recent data points we've seen -- up to and including January, since we get very real-time data -- is that 2016 mileage is not dramatic -- or not even different than 2015 mileage. And we've seen that even in the last data points of 2015 over 2014.

So the incremental -- yes, there was incremental, first half of 2015. That was a very measurable, quite notable, 3% to 5% on a month-over-month -- month-to-month basis, but we're starting to see that level out. We still see in the mix of miles driven, a little shift, about a 1% shift still, to trips over 15 miles. We are not seeing anything dramatic in seasonally adjusted braking rates; probably not a statistic most other companies talk about, but something that we're particularly interested in.

So, net-net, the gas price change may have played its way through. I'm going to say that with some ability to retract my statement next time we talk in May or whatever. So, stay tuned. But I would just correct that maybe the rest of the industry didn't see the frequency quite as high as perhaps an Allstate and Geico -- I think are the two that come to mind when we see that -- and certainly the two that took rate that was more reflective of that.

Glenn, all the miles driven trends you were just talking about, that's based on Progressive's own data?

Yes, we use the Snapshot population to see if we can get -- we're kind of nutty about this. We even look at it daily, weekly, whatever frequency we can. Because these are the sorts of things that for years we have reported, even to you -- we don't always know what drives frequency. You don't always know what drives severity. Well, now we actually do have at least one proxy for frequency, and miles driven is something we can measure. So it's only a piece of the puzzle but we're pretty obsessed with trying to get it.

I can imagine that. The only other thing I wanted to touch base on with regard to -- what some large writers of the industry in personal auto are seeing with regard to deteriorating margins. So, if the frequency impact that Progressive is seeing is not inconsistent with that of the industry, why do you feel that Progressive's personal auto margins haven't been really impacted at all?

Well, that's probably a longer story, but let me suggest to you that -- and we've done this on many, many calls. I've always said when we see situations that make us believe that rates will go up, we tend to take it early. It wasn't that many years ago that we took -- and talking in aggregates is not always the best thing, but it might work for this -- we took rates up; and, frankly, some of the trends didn't materialize to the degree that we thought. Our penalty there is that we actually grow little slower, with healthy margins.

However, in more recent times, we've gotten to a rate level that we thought was appropriate. And perhaps others weren't quite at that level, and therefore needed to take their rates up more. So the starting point matters. And that was my point before on -- I'd like to think that we're trying to ride the crest of a breaking wave. We want to stay on the crest all the time.

When you fall over the crest and your margins start to deteriorate, typically that's associated with a big hammer of rate. And we all know the implications of that to retention, to NPS, and to growth. The trick for us is really staying on that crest. And as I had mentioned earlier, as opposed to looking at rate take by rate filings, which you could look in a premium weighted adjusted basis, you'll see that the industry is about to 2.5% to 3.5% last year; some exceptions to that, for sure, on the high side and low side.

But I would always draw your attention to what is the realized average premium. And you see, in our case, that was about 4% in both channels. We have ways that I'm not going to get into in great detail of making sure that we build in some trend into our rating factors, as well, so that doesn't always come through in the published rate changes. So our objective is really stay on the crest of the wave.

And in answering your question, it depends on the starting point. So we were not inadequate, and didn't need to take a big rate increase. And we actually have been thinking small bites of the apple. And we're going to keep taking bites of the apple. My expectation is in the first quarter this year, we will take about -- probably the better part of 1 point, at least in our direct channel, and maybe close to that in our agency channel. So we will always take that kind of a rate.

I'm not going to speculate for the whole year. But it might be slightly below the 4%. Ian asked a question before about severities. And while we did have the fortuitous situation of severities going down, and therefore we could release reserves based on the negative severity trends we saw, I will tell you one of the first lessons I ever learned in this business was people don't get less expensive to fix.

So, we certainly don't plan negative trends into our pricing indications, and we're much more likely to reflect the kinds of numbers that Gary talked about going forward. So, long way of saying, but our job is to stay on the crest of the wave. And the good news is that we are not currently in the position of taking a significant hammer to our rates, or haven't had to for some time. And if we can stay on that crest of the wave without falling over, by taking small bites of the apple, that's our objective, and that produces some very nice outcomes.

That's great. Thank you.

This is Julia. I just want to remind everyone that if you'd like to ask a question, you may press star 1 on your touchtone phone. And while we wait and see if anyone else is calling in or has a question, I just wanted to make a quick housekeeping announcement which is that for those of you who would like to join us at our Investor Day this year, we will be holding it on the afternoon of Thursday, October 6, in Mayfield Village, Ohio.

And it looks like there are no further questions.

So, Shirley, I will turn it back over to you for the closing script.

Thank you. That does conclude The Progressive Corporation's Investor Relations conference call. And instant replay of the call will be available through Friday, March 18, by calling 1-866-501-0069, or can be accessed via the Investor Relations section of Progressive's website for the next year.