Globe Life Inc (GL) — Q2 2015 Earnings Call Transcript

Thank you. Good morning, everyone. Joining the call today are Gary Coleman and Larry Hutchison, our Co-Chief Executive Officers; Frank Svoboda, our Chief Financial Officer; and Brian Mitchell, our General Counsel. Some of our comments or answers to your questions may contain forward-looking statements that are provided for general purposes only. Accordingly, please refer to our 2014 10-K and any subsequent Forms 10-Q on file with the SEC. Then, I will now turn the call over to Gary Coleman.

Thank you, Mike, and good morning, everyone. Net operating income for the second quarter was $133 million or $1.05 per share, a per share increase of 30% from a year ago. Net income for the quarter was $127 million or $1 per share, a 2% decrease on a per share basis. With fixed maturities and amortized cost, our return on equity as of June 30 was 14.7% and our book value per share was $28.91, a 7% increase from a year ago. On a GAAP reported basis, with fixed maturities at market value, book value per share was $33.94, approximately the same as a year ago. In our life insurance operations, premium revenue grew 5.7% to $520 million while life underwriting margin was $139 million, down 1% from a year ago. Despite the growth in premium, underwriting margin declined primarily due to higher claims in Direct Response. For the full year, we expect life underwriting margin to increase 1% to 3% over 2014. Life sales increased 6% to $108 million. On the health side, premium revenue grew 8% to $232 million and health underwriting margin grew 4% to $52 million. The growth in underwriting margin lagged the growth in premium due to the large amount of group business added in 2014 which has lower margins than our other health business. For the full year, we expect health underwriting margin to increase 2% to 4%. Health sales increased 8% to $31 million. Administrative expenses were $47 million for the quarter, up 3% from a year ago and in line with our expectations. As a percentage of premium, administrative expenses were 5.7%, the same as a year ago. For the full year, we anticipate that administrative expenses will be up around 6% to 7% and around 5.8% of premium. I will now turn the call over to Larry Hutchison for his comments on the marketing operations.

Thank you, Gary. I will now go over the results for each company. At American Income life, premiums were up 9% to $207 million and life underwriting margin was up 6% to $64 million. Net life sales were $50 million, up 13% due primarily to increased agent counts. The average agent count for the second quarter was 6,603, up 15% over a year ago and up 5% from the first quarter. The producing agent count at the end of the second quarter was 6,516. We expect life sales growth for the full year 2015 to be within the range of 11% to 13%. Our Direct Response operations at Globe Life, life premiums were up 7% to $188 million, but life underwriting margin declined 16% to $37 million. Net life sales were flat at $45 million. We expect 4% to 6% life sales growth for the full year 2015. At Liberty National, life premiums were $68 million, approximately the same as a year ago. Our life underwriting margin was $18 million, down 3% from the year ago quarter. Net life sales grew 6% to $9 million. Our net health sales increased 4% to $4 million. The average producing agent count for the second quarter was 1,550, up 4% from a year ago and up 6% from the first quarter. The producing agent count at Liberty National ended the quarter at 1,550. Life net sales growth is expected to be within a range of 5% to 7% for the full year 2015. Health net sales growth is expected to be within a range of 2% to 4% for the full year 2015. At Family Heritage, health premiums increased to 8% to $55 million while health underwriting margin increased 5% to $11 million. Health net sales were up 4% to $13 million. The average producing agent count for the second quarter was 960, up 27% from a year ago and up 22% from the first quarter. The producing agent count at the end of the quarter was 969. We expect health sales growth to be within a range of 8% to 10% for the full year 2015. At United American General Agency, health premiums increased 16% to $88 million. Net health sales increased from $9 million to $10 million. Individual sales grew 21% to $7 million while group sales declined 6% to $2.8 million. For the full year 2015, we expect growth in individual sales to be around 15% to 20%. As we discussed last quarter, we expect lower group sales in 2015 due to the unusual number of large group cases we acquired in 2014. Premium revenue from Medicare Part D declined 11% to $75 million, while the underwriting margin declined from $9 million to $5 million. The decline in underwriting margin was in line with our expectations, due to the increase in Part D drug cost discussed on our previous call. We expect Part D premiums of $305 million to $315 million for the full year 2015. Expect margin as a percentage of premium to be approximately 6% to 8%. I'll now turn the call back to Gary.

I will spend a few minutes discussing our investment operations. First, excess investment income. Excess investment income, which we define as net investment income, less required interest on policy liabilities and debt, was $57 million, approximately the same as the second quarter of 2014. On a per share basis reflecting the impact of our share repurchase program, excess investment income increased 5%. We have discussed on previous calls the effect of Part D on excess investment income. Excess investment income was negatively impacted by Part D to the extent of $2 million in the second quarter of 2015. Excluding the negative impact of Part D, excess investment income would have been up 2% compared to the year ago quarter and up about 7% on a per share basis. For the full year 2015, we expect excess investment income to decline by about 1% to 2%; however, on a per share basis we should see an increase of about 3% to 4%. At the midpoint of our 2015 guidance, we're expecting a drag on excess investment income from Part D of approximately $8 million. Now regarding the investment portfolio, invested assets were $13.6 billion, including $13.1 billion of fixed maturities and amortized cost. As of the fixed maturities, $12.5 billion are investment grade with an average rating of A- and below investment grade bonds are $580 million compared to $563 million a year ago. The percentage of below investment grade bonds to fixed maturities is 4.4%, the same as a year ago. With the portfolio leverage of 3.6 times, the percentage of below investment grade bonds to equity, excluding net unrealized gains on fixed maturities, is 16%. Overall, the total portfolio is rated A-, same as the year ago. In addition, we have net unrealized gains in a fixed maturity portfolio of $1 billion, approximately $935 million lower than at the end of the first quarter. The decline in unrealized gains was generated by higher interest rates, not by concerns over credit quality. Due to the recent events in Greece, I'd like to assure everyone of our limited exposure there. We have no direct exposure to Greek sovereign debt and we have no exposure to companies that do business primarily in Greece. We don’t expect to realize any losses should Greece exit the Eurozone. To complete the investment portfolio discussion, I'd like to address our investments in the energy sector. I believe the risk of realizing any losses in the foreseeable future is minimal for the following reasons. Over 96% of our energy holdings are investment grade. At the end of the second quarter, our energy portfolio had net unrealized gains of $69 million. Less than 8% of our energy holdings are in the oilfield service and drilling sector. We have reviewed our energy holdings and concluded that while we may see some downgrades, we believe that the companies we invest in can withstand low prices for an extended duration. Now to investment yield; in the second quarter we invested $250 million in investment grade fixed maturities primarily in the industrial and financial sectors. We invested at an average yield of 4.7%, an average rating of A- and an average life of 30 years. For the entire portfolio second quarter yield was 5.85%, down 7 basis points from the 5.92% yield in the second quarter of 2014. At June 30, the portfolio yield was approximately 5.83%. The midpoint of our guidance for 2015 is the same as the new money yield of 5.0% for the two quarters of the year. And one last thing. We are encouraged by the potential for higher interest rates. As discussed previously on analyst calls, rising new money rates will have a positive impact on operating income by driving up excess investment income. We are not concerned about potential unrealized losses that are interest rate driven reflecting on the balance sheet as we would not expect to convert them to realized losses. We have the intent and, more importantly, the ability to hold our investments through maturity. Now I'll turn the call over to Frank to discuss share repurchases and capital.

Thanks, Gary. First, I'd like to briefly discuss a few items impacting our 2015 earnings guidance. As Gary mentioned, growth in life underwriting income lagged behind the growth in premium in the second quarter due to higher policy obligations in our Direct Response operations. In the second quarter this year, policy obligations at Direct Response were 52% of premiums versus 49.1% in the first quarter and 48.1% for all of 2014. As discussed on our last call, we thought the percentage would trend higher during 2015 and be around 49% for the year primarily due to anticipated higher claims related to policies issued in calendar years 2000 through 2007. With claims data through June 30, we are now seeing higher claims than anticipated on policies issued in 2011 through 2013 as these policies exit a two-year contestability period. Beginning in 2011, we introduced the use of prescription drug database information into our underwriting procedures for certain adult policies with an expectation that our mortality experience would be better on such policies than historical experience. While actual mortality related to policies issued in 2011 through 2013 has not been greater than historical levels, they are higher than we assumed when the policies were issued. Approximately, 9% of the premium collected in 2015 relate to policies issued in 2011 through 2013 that used the prescription drug database. We believe the higher than anticipated claims will continue throughout the year and, thus, we are now revising our estimate of policy obligations for the full-year 2015 to a range of 50% to 51% of premiums. At the midpoint of this range, the Direct Response obligations will be approximately $12 million higher than previously estimated. This increase in the expected policy obligations at Direct Response is the primary driver of the $0.05 reduction in the midpoint of our guidance from $4.28 to $4.23. Now regarding our share repurchases and capital position. In the second quarter, we spent $86.3 million to buy 1.5 million Torchmark shares at an average price of $56.93. So far in July, we have used $15.8 million to purchase 269,000 shares. For the full year through today, we have spent approximately $192 million of parent company cash to acquire 3.5 million shares at an average price of $55.25. The parent started the year with liquid assets of $57 million. In addition to these liquid assets, the parent will generate additional free cash flow during the remainder of 2015. Free cash flow results primarily from the dividends received by the parent from the subsidiaries less the interest paid on debt and the dividends paid to Torchmark shareholders. We expect free cash flow in 2015 to be in the range of $355 million to $360 million. Thus, including the $57 million available from assets on hand at the beginning of the year, we currently expect to have around $417 million of cash and liquid assets available to the parent during the year. As previously noted, to-date, we have used $192 million of this cash to buy 3.5 million Torchmark shares, leaving around $225 million of cash and other liquid assets available for the remainder of the year. As noted before, we will use our cash as efficiently as possible. If market conditions are favorable, we expect that share repurchases will continue to be a primary use of those funds. We also expect to retain approximately $50 million to $60 million of liquid assets as a parent company. Regarding RBC at our insurance subsidiaries, we plan to maintain our capital at the level necessary to retain our current ratings. For the last two years, that level has been around an NAIC RBC ratio of 325% on a consolidated basis. This ratio is lower than some peer companies but it is sufficient for our company in light of our consistent statutory earnings, the relatively lower risk of our policy liabilities and our ratings. As of December 31, 2014, our consolidated RBC was 327%. We do not anticipate any significant changes to our targeted RBC levels in 2015. As we have discussed on prior calls, S&P changed their view last year after the treatment of certain intercompany preferred stock and requested additional capital be contributed to our insurance subsidiaries to retain our credit ratings. We have reviewed various alternatives available to us and are scheduled to meet with S&P in August or September where we will discuss potential solutions and courses of actions with them. Based on our analysis to-date, should we decide to add additional capital, we believe we will be able to address the additional capital needs without significantly impacting our free cash flow available for buyback. One option available is for Torchmark to issue additional hybrid securities treated as debt for financial reporting purposes, but equity for S&P capital purposes. If we were to issue such securities in an amount sufficient to meet the entire shortfall, we estimate that the overall impact on EPS would be less than $0.01 per share. Those are my comments. I will now turn the call back to Larry.

Thank you, Frank. For 2015, we expect our net operating income to be within a range of $4.18 per share to $4.28 per share, a 5% increase over 2014 at the midpoint. Those are our comments. We will now open the call up for questions.

Hi. First, I just have a question on the response claims and you get the amount and the impact on the benefits ratio. Seems like the amount on an annual basis should be about $0.06 a year, so if you could confirm whether that is right. And then, should we expect that will continue into next year and at least for the next few years? And then, secondly, on the agent count at American Income has dropped from beginning to ending obviously on an average basis it was up, maybe you can discuss what drove the decline and what your expectations are for agent count growth at American Income?

Jimmy, you are correct about the $0.06 impact for 2015. As for the overall additional impact, we anticipate that the policy obligations will be in the range of 50% to 51% for 2015. Looking ahead to 2016, our best estimate suggests that the policy obligation for Direct Response may rise to about 51% to 52%, with the Direct Response margin potentially decreasing to around 20% to 21%. This outlook is based on the data we currently have and our projections for the future.

Okay. And on the agent count American income?

And just to be clear on the Direct Response business, it's not that you have seen a sudden spike in claims but it's more you would assume that claims get better and the margins would be better because of the use of prescription drug information, and in reality they just have not been, right?

Jimmy, that is exactly correct.

And this decline this quarter was primarily due to one specific block rather than being distributed across various subsidiaries or other parts of the business. Are those policies related to that specific block?

Yes, largely that is correct, I mean there is a little fluctuation, we had some seasonal fluctuation but largely the case.

And those are like the fluctuations are just normal volatility in claims from quarter-to-quarter, right?

That is correct.

Okay, thanks. And then lastly just on the agent count drop at American Income it did grow on an average basis but it was down from the end of the previous quarter. So maybe just if you could discuss what drove that and your expectations growth at American Income?

Jimmy, this is Larry. Meeting agent count is less important to the average agent count, there's some fluctuations every quarter just on the last day it depends on the terminations that comes through. If you look at the overall results for the last year we see we have some agent growth. Now we still expect to meet our producing agent count projection of 6,800 to 7,000 agents for 2015.

Okay. Thank you.

Hi, thank you. I just had one follow up first on Direct Response. Since you now identified two blocks of issue or the policies from 2000 through 2007 and then the 2011 through 2013, and so I guess was there any difference in kind of your underwriting or pricing assumptions from kind of that 2007 through 2011 period that gives you comfort that you won’t see any higher incidence of claims there? And I guess the same question would be for 2014, 2015. Were there any changes that you made to your unit pricing or to your assumptions for the most recent years?

Yes, Erik. The change from 2011 to 2013 primarily involved lowering the overall mortality assumptions due to our use of the prescription database. This assumption continued through the issue years of 2014 and 2015, which are still in the contestability phase, so we haven't seen any claims from those yet. We are gradually adjusting how we utilize the Rx data, and we will examine why we are not experiencing the anticipated benefits, making necessary decisions for 2016. We believe this will remain relevant to the block from 2011 to 2015. However, it differs from the earlier period of 2000 to 2007, where we observed higher mortality rates, which have been factored into our overall assumptions for the later years.

Got it. So there was a change in your assumptions kind of in the 2008 period?

Really, yes, overall with regard to some of those earlier years.

Got it. Okay. So you don’t expect the issues you are seeing in the 2000 through 2007 block to continue into the next few years?

That is correct.

Okay, thank you. And then just one question you gave the target you had for the year-end the agent count for American Income. Would you mind providing any update for Liberty National as well as Family Heritage given the strength that you have seen in recruiting them in past couple of quarters there?

Sure. We expect the year-end agent count in Liberty National to be in a range of 1,630 to 1,660 agents. At Family Heritage, we expect the year-end agents count to be in a range of 975 to 1,000 agents.

Good morning. I want to go back to direct response business if I could, and couple of questions there. One is on the previous call I think you talked about the early 2000 vintages being the ones that showed claims activity, now you are talking about 2007. So does that suggest that you have seen elevated claims activity now really move a little further to the newer vintages as well beyond the 2011 to 2013 issue that we discussed?

Good morning. I want to revisit the direct response business and ask a couple of questions. Previously, you mentioned that the early 2000 vintages were the ones showing claims activity, but now you're discussing 2007. Does this indicate that you've observed increased claims activity moving further into the newer vintages, beyond the 2011 to 2013 issues we talked about?

Okay. I just want to clarify. In the last call, you mentioned the early 2000 vintages, which are over a decade old. Now, you're discussing the years 2000 to 2007. Does that mean the years 2005, 2006, and 2007 also fall into that category?

Now, when I discussed the issue years from 2000 to 2007, that is the same vintage we referred to in prior calls.

Okay.

We really haven't changed our outlook right now with respect to additional claims on that particular block.

Okay. And had the 2011 to 2013 vintage data not developed the way it had, will you still have expected the benefits for this year to fall and within the 48.5% to 49% range, which you've previously offered?

Yes, been really close to that 49%.

Okay. And maybe one last question on this direct response business. How quickly do you expect the 2000 to 2007 and the 2011 to 2013 vintages to run off? What's the rate of the decay here?

Yeah. I'm not sure. I mean, the obviously run off was over really long period of time.

Yes.

What we see right now is that the peak of the adverse experience is expected to occur around 2017 when the 2015 years consolidate their contestability period. We view this as the low point for direct response margins, after which we anticipate improvement.

Okay. And I'm sorry. Maybe I sink in one last one. In direct response, we're also seeing a bit of a slowdown in sales. Is that just distributable to repricing of that business now that mortality data has come in a little higher than expected?

I think if you recall, the second quarter of 2014 was the largest projection quarter in the history of direct response. So actually, we're pleased with the slight increase this quarter. We still expect an increase in sales in direct response this year in the range of 4% to 6%.

Yeah. Good morning everybody. I wanted to ask a bit more about the direct response. Frank, did the years 2000 to 2007 perform in line with your current expectations for the quarter?

For the quarter, we saw just a little bit higher seasonal fluctuation than we really do anticipate coming back to the normal trend over the course of the year. Our expectation for the full-year is still in that kind of in that range we talked about last time, probably increasing the overall obligation percentage by 0.4%, 0.5%.

Okay. Great.

And so we really haven't changed our overall outlook for that.

Okay. Great. And then on the newer stuff, could you explain the importance of the contestability period ending in this calculation or how you see things? And why is that so important?

Sure. Well, for the first two years after issue we have the ability to contest any claims that come in during that period of time.

All right.

After the two-year contestability period ends, the claims become non-contestable unless we can demonstrate certain conditions related to the application. Historically, the third year tends to have the highest number of claims, which then decreases afterward. This is when we begin to see those early claims.

Okay. Is that just timing? Anyway, all right.

Yes, it is just the timing of that.

Steven, I would add that Frank has mentioned 2011. We didn’t start to use the Rx information until late in 2011. So really 2012 is the first year we really had enough issues where can start to see and experience in late 2014.

I'm going to stick with that topic, because after Schwartz asked those questions, I guess I'm not clear. This type of direct coverage, would you describe this as final needs-type coverage? And the reason I ask is that it seems like you're having mortality events relatively quickly. Is that the right way to characterize this type of coverage?

In general, yes. And those tend to be very quick.

And so when you all said that you were exiting the contestability period, what is it that you have been successful on disputing in that period, and does that have anything to do with the Rx data?

I'm sorry, Randy, on that, you kind of cut out on that little bit. I didn't quite catch that whole question.

In the contestability period, what is it that you were contesting, and does that have anything to do with the Rx data or is it more typical contestability type stuff?

It's more typical contestability issues. You're assessing the answers and information provided to you and checking for any misrepresentations regarding the application. We use the Rx data to confirm that we have authorization from them and to verify whether the information in the application is accurate.

Some of the blame that Rx is not in the contestability period, but it's a time issue. You have a better underwriting picture and so you either decline some of the business you otherwise would have issued, or some of that is rated as substandard business. So it's not really just a contestability period, it's evaluating the risk in your underwriting for the life insurance.

Really not

We didn't see that much difference during the contestable periods for these claims. And remember, the issue here is not that the mortality is worse and what we experience in the past, what has happened is we've experienced about the same mortality as we did before we started using the prescription drug as an agent. But the problem was we assumed that we were going to have better mortality in our reserves and that’s why you're seeing the increase in the policy obligations. That’s an overall. What we need to look at is the Rx in certain segment that we think probably is benefiting and others is not, and we will have to evaluate what it does and determine how we use that going forward. But I do want to emphasize, we're not seeing worse mortality than we saw before. We're seeing about the same. The problem is, we thought the Rx would lead us to better mortality.

Just a quick question on follow-up to Mark Hughes. If the receivable stays constant on the recovery from the government, wouldn't it be a neutral next year on investment income? At some point, this reverses. But if you reverse it, once you got up and put new stuff up, it seems like it would be a neutral to investment income.

Bob, you're right it would be neutral. It would be about the same drag next year as this year.

Okay. And at some point, it would reverse, right? Do you have a sense on what year that would be?

It should reverse by the end of 2016. And then of course depending upon what happens with 2016 claims activity and the receivables and whatever is generating new in 2016.

Right. So if the drag stays the same, but it's not an incremental drag, so investment income should move up with cash flow and yields and not be impacted in 2016, and then it becomes an equivalent positive in 2017 to the negative it's been in this year?

Yes, that's correct. And when I had answered it, I was looking at just a drag, not an incremental drag, but it's correct, it would be the similar drag in '16 as it is in '15 but then presuming that the receivables actually get to go down or by the end of 2016 you would see the real incremental benefit in '17.

Okay. Buyback. Last year's annual report, I think you said you were getting near, but hadn't reached intrinsic value where buyback was the first best use. But you sort of sent a warning that if the stock kept running, last year it was up 4%, it's up, even with the correction today, 10%. Are we anywhere near the point where the warning has to be sent out that dividends might be a use, or is buyback still below intrinsic value?

Bob, we still believe that the buyback value is below intrinsic value and we are currently trading at a higher multiple, even compared to when we wrote the annual report. However, we maintain that we haven't yet reached intrinsic value. Therefore, we will continue to buy, as the returns we are seeing significantly exceed our cost of capital and also surpass what we could achieve through alternative uses. We will keep this strategy in place. As we've mentioned before, if the stock price reaches or surpasses intrinsic value, we will reevaluate our approach.

I guess, first one, I wanted to follow-up on the prescription drug data. I guess just to be clear, do you only use that when you price Direct Response business, or did you also use that in some of your other businesses when you were underwriting those?

Yes, it has been used in just very limited situations with respect to some older age issuances in the other agencies. And again I will stress it's very limited circumstances and we had not reduced any of our mortality assumptions for the use of Rx in those other agencies. So it's just simply been just an added tool in the underwriting process there.

So it's really isolated to Direct Response at this point, for the most part, is that correct?

That's correct.

Okay. And then just follow-up to Randy's question on the RBC changes. I know it's early on still, but it seems like the rating agencies tend to use higher capital charges than the RBC formula already. So I mean is it your best guess that even though RBC ratios will change and go down for the industry, that it won't necessarily change the way that you and others are managing capital?

I think that is a very real possibility and you're right S&P has their own capital factors that they use, which are higher, and would really be more similar to what the NAIC is looking to move towards, and then Moody's and A.M. Best how they would look at it. But we would anticipate that, or at least, it would be a possibility that no changes at all would be necessary.

Thank you. Just to come back, one more thing on this Rx issue. It seems to me, if I'm understanding what you're saying, when you went to that, you assumed mortality would improve. And so I would presume that had some impact on your pricing decisions. Now that it hasn't, it would suggest, I guess, that you're underpriced. How much are you thinking right now you may need to raise prices, excuse me, if the Rx data just isn't giving you what you need?

It's too early to provide a definitive answer to that. We will be examining the situation, but it doesn't mean we will necessarily have to increase prices. It may only apply to specific segments, age groups, or circulation levels where we might need to raise prices, or we might decide not to continue selling in those areas. We need to conduct more analysis before determining whether to raise prices or discontinue certain segments.

Okay. And then a second question. In looking at the premium growth this quarter, not only was it, I think, the strongest we've seen in over 10 years on the life side, but also on the supplemental side. And has some of your strategy changed there because of growing this up beyond Family Heritage, and really how much longer do you think you can keep this growth rate going? Because I think you've probably got about the strongest organic growth rate in the industry today.

On the life insurance side, we have observed higher premium growth and believe we can maintain this momentum. In American Income, which represents our largest business segment, we are achieving growth around 9% and anticipate continuing this trend. Additionally, in direct response, our second largest segment, we are seeing growth exceeding 5%. We are confident we can sustain at least the 9% growth level. Similarly, on the health insurance side, we are optimistic about maintaining premium growth, especially since we had experienced declines in health premiums a couple of years ago after exiting certain health blocks. Overall, we feel positive about the future regarding premium growth.

What about the general agency this quarter on the supplemental side? It seems to be surprisingly strong.

We've seen solid growth in our individual sales, which has been stronger this year compared to last. This is contributing positively to our performance.

Okay, and then the final one. Again this quarter further improvement in persistency, particularly thinking on the renewal year. How much stronger is that now than what you're pricing for? And what does this say really about your underlying core earnings growth rate since I would that is a significant benefit?

Colin, I'm not sure I can answer how that's different from what we're repricing. I know we've seen improvements over a lot of products. I just quantify here on the call.

Hi. First question is to do, let's want to make sure I've this right, did you say 9% of total in-force direct response block was the Rx related underwriting was that the right quantification?

That's correct.

It was the premium received on the '11 to '13 years actually. Just to clarify.

Okay. So how do we think about, of your new sales this year so far, how much are Rx, using that Rx data? Can you quantify that? Is it 50%? Is it 100% of Direct Response sales that are relying upon this data?

Yes, it's around 50%.

Okay.

50% of 2015 sales would be going out using the Rx.

Got you. Okay. So it's possible you are going to be repricing 50% of your sales for Direct Response, or is that not the right way to think about it? Is it somehow isolated that the problematic parts are not the entirety of the 50%? How do we think about that?

That's correct. As Gary and Larry mentioned earlier, we still need to finish determining which segments we are not benefiting from and where that falls within the 50%. Some portion of that might be bringing us some incremental benefit, but that’s where we really need to focus our efforts, so it shouldn't be a significant issue.

Okay. So it's going to be some portion of that 50% of total sales?

This is Larry. It's important to note that the prescriptions in 2011 were less advanced than those in later years. As models are developed to gather information, we can identify better combinations of medications that reflect health history. Therefore, you shouldn't expect the same outcomes from 2011 and 2012 as you would from 2013 and 2014. We need to allow for some of these facts to unfold, and we are currently in that process. In fact, the experiences of 2013 and 2014 may differ from those of 2011 and 2012.

Understood. And just to put this in context, when you look at the block, if you will, that you have identified thus far that you deem to be underpriced, are we talking about a block that's actually losing money? Is it just subpar returns? Can you provide some context around that?

Yes, Tom, we're definitely not losing money. We're pressuring the margins. Overall margins are being pressured. We've been in the 23%, 25% underwriting margin for Direct Response over the last few years. This year is going to be closer to 21%. And it's early, and our preliminary estimates of how this plays out, we don't see that profit margin going below 18%, 19% at the worst case. So even if that all develops on that basis, we still are going to have 18% to 19% profit margin. We're not in a position of losing money at all. It's just that margin is not as high as it has been in the past.

So the lowering of that range could be higher than the 18% or 19%.

Yes, the low end is 18% and 19%. That could be the worst case, so it may be somewhere between that and 21%. Over time, as we make pricing adjustments, we will improve.

Okay. And then my last question on that is, when you look back to when you began to use the Rx data and pricing on that basis, was it done in response to the market becoming a lot more competitive for you? Had it become more price elastic than it was historically? Like what was the driver of starting to use this, and has that overall part of your business become more price sensitive?

In 2011, we viewed this as a tool to more accurately assess and evaluate the risks we would be underwriting. However, we believed it would have a more beneficial impact than it ultimately did, which was an error in judgment. The actual experience has not been as profitable as we expected, influencing our marketing strategies in the lower performing segments. I don't think this comes as a major surprise; the business remains profitable, but we had hoped for a higher profit level in 2011 than what has materialized.

We have always considered the response in Direct Response because we can't conduct extensive underwriting due to time and call limitations. This has been the historical approach. When we learned about the use of prescription drugs, as Larry mentioned, we saw it as an affordable way to gain better information for underwriting. If we could improve our underwriting, we might be able to explore segments we hadn't considered before. The intention was not to compete with others but to enhance our underwriting process.

Hey, thanks for extending the call for a moment for a quick question or two. Just following up a little bit on Tom's question, if we think about Direct Response overall, maybe it was a low to mid-20s margin, and maybe it's got downside for one or two particular years, down to the very high teens. So if we call it about a five-point swing in that margin, can you translate that to ROE of the business?

We don't really calculate an ROE on the business. Yes, our return on investment might be lower, but we haven't really calculated that, and I don't have an answer for that.

Now, and John, do you mean for that business as a whole or just the overall ROE for the impact that would have on Torchmark's overall ROE?

Could you provide more details on whether this is related to the Direct Response life business or the overall effect on Torchmark? Clearly, the total impact on Torchmark would be much smaller.

Yes, I mean that's what you probably, looking at a half a percent or something to that effect there I would guess. But I agree with Gary as far as looking at the business. We don't really look at it in that way or don't have that any way right now.

Thanks very much. Two questions related to Direct Response. Do I have it right when I say that with respect to the 2000 to 2007 block, that in contrast to the 2011 to 2013 block, in which you are not experiencing higher than expected mortality, you're just not getting the improvement that you had anticipated, in the earlier block you are experiencing higher than expected mortality?

That is correct.

Is that right?

That's right.

It appears that when an insurance company faces a higher than anticipated number of death claims or larger claims, it could be due to various factors. Based on your analysis of these claims from the seven issuance years, what is your initial or best understanding of the underlying issue?

As we've really taken a look at those claims yes, interestingly enough there really isn't one particular area that seems to be sticking out, if you will, as far as where those additional claims might be coming from. The only thing that that tends to may be a little bit higher than what would be normal average would be some deficit as relating to respiratory illnesses. Other than that that there really is not, when we looked at how we segmented in different areas really it's very little that sticks out.

Good morning. I want to go back to direct response business if I could, and couple of questions there. One is on the previous call I think you talked about the early 2000 vintages being the ones that showed claims activity, now you are talking about 2007. So does that suggest that you have seen elevated claims activity now really move a little further to the newer vintages as well beyond the 2011 to 2013 issue that we discussed?

Yaron, I’d like to clarify or maybe make sure I clarify here. We have a good handle on this and still expect those claims or the earlier discussions on 2000 through 2007 to interact with what we expect in the 2011 through 2013.

Okay.

So it's really more the timing for us of the claims as far as how we're growing in these spaces.

Yes, and as far as on the higher mortality we do expect that. As we get to the point that this part of the business will stabilize we do expect it to be more in line or your best assessment as we gauge those outcomes. The severity has been a little higher than we expect prior years. So overall, the expectations that we have there in those earlier vintages, we would anticipate will drop to normal over a period of time.

Yes. Thank you for taking my question. My question relates to the growth in agent count. In your earlier comments, I believe you mentioned that the agent count at American Income was up 15% year-over-year. What is driving that growth? How much of it is agent recruitment, and how much of it is increased productivity?

Thank you for the question. That increase in agent count largely has been driven through an increase in productivity but we've also had some very strong recruiting quarters which have helped set, too. And we’re doing this across multiple channels. So the combination of both has led to the increase.