Hi, good afternoon, everybody. For those of you who have not met me, I'm Seth Weiss. I took over the life coverage in December. You're not here to see me. I'm going to introduce Gary Coleman and Larry Hutchison, co-CEOs of Torchmark. Torchmark focuses on providing protection-oriented life and supplemental health insurance to the low to middle class. It yields very strong free cash flows and is consistently at or near the top of the life group in terms of cash payout ratios. Gary and Larry took over the CEO role last June. Prior to their current roles, Gary served as CFO and Larry as general counsel. Each has over 25 years of experience with the companies. With that, I'll turn over the microphone.
Thank you, Seth. Good afternoon. The mics are bigger than we're used to. Before we get started, please be aware our presentation includes forward-looking statements. Please take note of this first slide.
We don't have a slide though. There we go. There we go. Okay. As Seth mentioned, Torchmark had a leadership transition last year when Larry and I became co-CEOs. We're often asked why Torchmark chose the co-CEO structure. While we admit it's a little unusual, we think it's the best arrangement for the company because of the unique relationship that Larry and I have. We've worked together for over 26 years at Torchmark. We've worked in all of our insurance operating subsidiaries. Through that experience, our complementary skill sets, primarily our trust in each other, we think this arrangement provides a depth to the position that wouldn't exist with a single CEO.
The co-CEO arrangement facilitates a sound collective decision-making process. It also allows us the flexibility to more easily oversee daily operations and plan for the future. We both believe in the Torchmark business model. We sell simple protection-oriented products to middle-income families through captive distribution with products that generate strong margins and high cash flow. We maintain a conservative investment approach. We manage our capital to maximize shareholder value. Our target market is vastly underserved. It provides significant opportunity for growth, as has been evidenced by the company's performance in recent years.
Due to our focus on stable, protection-oriented products and the conservative management of investments, Torchmark Corporation has delivered consistent growth. On this slide, we show operating earnings per share. I might mention that in 2012, life insurance companies adopted accounting standard ASU 2010-26. It's the new accounting standard regarding the deferral of acquisition costs. As part of that adoption, life insurance companies restated earnings for the previous five years for comparability. On this slide, the EPS numbers for 2007 to 2012 reflect the new standard, while the years are noted by the asterisks, 2003 through 2006, reflect the previous standards. As you can see, under either standard, net operating earnings per share has increased at a compound annual growth rate of 8.3% over the past 10 years, despite the financial crisis.
For the last five years, earnings per share on the new accounting basis has grown at an 8.7% rate. For 2012, we grew net operating earnings per share by 15%, our best year in the past 14 years. Torchmark Corporation has also shown consistent growth in book value per share. Excluding net unrealized gains and losses on our fixed maturities, our book value per share grew at a compound annual rate of 9.7% over the last five years. While our reported GAAP book value, which includes the unrealized gains and losses, grew at a compound annual rate of 16% over the past five years. As I mentioned, 2012 was a very good year for Torchmark Corporation. Our highlights include, first of all, that the net operating earnings grew 15%. At American Income Life Insurance Company, life sales grew 12%, while the agent count increased 18%.
Our direct response life sales grew 3% despite a difficult economy. We also began a turnaround at Liberty National Life Insurance Company. Agent count and sales have grown steadily since February of last year, and the life underwriting margins increased from 22% of premium to 26%. We completed the acquisition of Family Heritage Life Insurance Company of America and did so without restricting the future buybacks or M&A activity. We refinanced quite a bit of our debt on favorable terms, and we repurchased 7.5 million shares or 7.4% of the outstanding Torchmark Corporation shares. All in all, it was a very good year for Torchmark Corporation. I'll ask Larry Hutchison to give his comments on our insurance operations.
As you can see, the largest component of net operating income is underwriting income. Underwriting income is about 70% of our pre-tax operating income. Let's look at a breakdown of underwriting income. As you can see, life underwriting income is the largest component of underwriting income. It produces about 72% of our underwriting income, while health, excluding Part D, produces 16%. We focus on life insurance as it has higher margins, generates significant investment income, is less competitive and less regulated than health insurance. Let's look at our major operating units. American Income Life Insurance Company is our largest and most profitable distribution channel. As you can see on this chart, life premiums at American Income Life Insurance Company have steadily increased at a 10-year compound annual growth rate of 9.1%. Underwriting margins as a percentage of premium have consistently ranged from 30%-33% before administrative expenses.
2012 was a great year for American Income. Life sales were up 12% over 2011, and agent count was up 18% over a year ago. We continue to focus on developing middle management to ensure sustainable agency growth. We also continue to refine our agent training programs and financial incentives. Our laptop sales presentation provides us with a wealth of valuable data we can use to manage the agency force. The data allows us to monitor and break down agent activity at all levels of the organization, by individual agent, by individual managers, and by region. American Income's labor union affinity and underserved target market allow it to operate in a niche that provides plentiful opportunity for future growth.
We expect to see strong net life sales growth throughout the year, beginning in the low to mid-single digit growth through the first quarter, then rising to 10%-14% for the entire year. Our second-largest distribution channel is direct response. As you can see on the chart here, direct response life premium has grown at a 10-year compound annual growth rate of 7.1%. Our basic strategy is to continually search for new ways to reach our market. We take an innovative approach that's constantly evolving. Over the past few years, internet marketing and our inbound call center have been our fastest-growing sources of new production. Five years ago, internet marketing and inbound calls produced about 5% of direct response new business. Today, internet marketing and inbound calls produce approximately 40% of our direct response new business. Despite a difficult economy, direct response life sales grew 3% in 2012.
We expect sales to be relatively flat in the first quarter. As I said in our conference call last week, we are optimistic that the initiatives we're putting in place throughout 2013 will increase response rates, resulting in mid-single-digit sales growth for the full year. I would like to add here that the post office or the Postal Service's recently announced plan to discontinue Saturday mail delivery is not expected to have any impact on our direct mail business. As you can see on this chart, Liberty National has a history of stagnant premium growth. Several years ago, we began to convert Liberty National from a fixed cost model to a variable cost model. We accelerated this process and began a turnaround program late in 2011.
We made a management change at Liberty and several other significant changes, including office operating expenses are now the responsibility of the branch managers rather than the home office. All new agents are hired as independent contractors rather than employees. We added a new layer of middle management. We cut a significant portion of our non-deferred acquisition costs. While this is a significant change in the culture of this agency, we knew the changes were necessary to produce acceptable profit margins on the business we write and to put Liberty in a position to grow going forward. The changes we've made are paying off as we had steady sequential growth in both sales and agent count in 2012, beginning in mid-February, and our life underwriting margin as a percent of premium has increased from 22%-26%. We're excited about the acquisition of Family Heritage.
This is the kind of company we've been looking to acquire, a company selling protection insurance to middle-income families through a captive agency. Their offering of return-of-premium health products in non-urban areas gives them a unique operating niche. The integration of Family Heritage's operation has gone very smoothly so far. We believe there's a potential for strong long-term sales growth through geographic expansion and integration of Torchmark's agent recruiting techniques. As we said earlier, this acquisition will not restrict our future share repurchases or M&A activity. I'll now ask Gary to discuss our investment operation.
The second major component of operating income is excess investment income. In 2012, it's $237 million, or about 30% of pre-tax operating income. Excess investment income is our net investment income, less the required interest on the net policy liabilities and the interest on our debt. The primary component is the investment income earned on our $12 billion investment portfolio. Regarding the interest on the policy liabilities, since we sell basic whole life and term insurance that is not interest sensitive, the interest on the policy liabilities is primarily the interest resulting from the discounting of the liabilities at the GAAP discount rates. Over the years, we have followed a conservative long-term investment strategy. With the high underwriting margins in our insurance products, we don't have to stretch for investment yield. As shown on this slide, 96% of our investment portfolio consists of fixed maturities.
These assets are primarily long-term investment-grade corporate bonds. We have no commercial mortgage-backed securities or securities backed by subprime or Alt-A mortgages. We invest in long-term fixed-rate assets because they provide the best match for our policy liabilities that have long duration and fixed rate nature of their own. Of the $12 billion of fixed maturities, $11.4 billion are investment grade and $585 million of bonds are below investment-grade bonds. The percentage of below investment-grade bonds to total fixed maturities is 4.9%, compared to 6.4% a year ago, and as you can see, higher levels in prior years. At the current level, and with a portfolio leverage of 3.5x, the percentage of below investment-grade bonds to equity, excluding net unrealized gains and losses, is 17%, which is less than most of our peers. Overall, the portfolio is rated A minus.
I would like to discuss the current interest rate environment, the topic that seems to be of most concern to investors in the life insurance industry. While the lower interest rates pressure investment income of all life companies, we believe that we face less exposure to a lower for longer environment than most of our peers. As long as we're in this low interest rate environment, the portfolio yield will continue to decline and thus pressure excess investment income. However, the impact on Torchmark will be diminished by the fact that on average, only 2%-3% of fixed maturities will run off each year over the next five years. To quantify the potential impact of an extended low rate environment, we performed a stress test assuming a new money rate of 4.25% for all investments made in the next five years.
This scenario results in a portfolio yield of about 5.55% at the end of 2017. If you run the test again, assuming a new money yield of 4%, it only makes about a 5 basis points difference, or in other words, the portfolio yield at the end of 2017 would be 5.50%. At these rates, we would still earn a small spread on the net policy liabilities while earning the full 550-555 basis points on our equity. In either scenario, we will still generate substantial excess investment income. In addition, unlike many of our peers, we don't have concerns regarding the potential impact of low interest rates on our benefit reserves or DAC. Companies that sell interest-sensitive life, variable life, and variable annuity policies follow the accounting standard that's previously referred to as FAS 97. That standard requires annual unlocking of assumptions used to calculate reserves in DAC.
In a lower for longer environment, companies may have to lower assumptions regarding future interest rates, resulting in immediate increases in reserves and/or write-downs of DAC. As mentioned, we sell simple whole life and term products. These products are accounted for under the accounting standard previously referred to as FAS 60. As such, DAC is amortized based on premiums earned, not gross profits. In addition, interest rates are locked in at policy issuance and are only changed if it is apparent that the policy is in a loss situation. Because of our high underwriting margins, it is unlikely that we would ever experience a loss recognition situation. For loss recognition to occur because of lower interest rates, our average portfolio yield would have to decline to 4% and remain there permanently. Our portfolio yield at the end of 2012 was 6.04%.
Even if new money rates remained at 4% for the next 20 years, our portfolio yield would still be in excess of 4% 20 years from now, thus no loss recognition. With respect to our statutory balance sheet, we perform cash flow testing each year for regulatory purposes, our statutory reserves are more than adequate. In fact, in the New York 7 cash flow testing performed in 2012, the margin of adequacy in each of the company's statutory reserves was substantial. In summary, with high underwriting margins, we generated significant underwriting income don't have to rely on excess investment income to generate positive earnings. However, from our stress testing, we are confident that we can maintain or grow the current level of excess investment income per share in an extended low rate environment. That said, we strongly prefer higher interest rates.
Because of our product profile our strong and consistent cash flow, we would greatly benefit from a spike in rates. We would not be concerned about the resulting unrealized losses in investment portfolio because we have the intent and more importantly, the ability to hold our investments to maturity. Now I'd like to move on to capital management. This slide shows the free cash flow generated at the parent company in each of the last 10 years. We define free cash flow as the cash that is available to the parent company from the annual dividends received from the subsidiaries, less the interest expense on our debt and less the dividends paid to Torchmark shareholders. The net amount left over is free cash that can be used for any corporate purpose.
Because of the products we offer and our high underwriting margins, we have a large, stable in-force block that consistently generates substantial free cash flow year after year. This graph shows our free cash flow over the past 10 years. As you can see, we generated strong free cash even at the height of the financial crisis. In 2013, we expect to generate free cash of around $355 million to $365 million. On an ongoing basis, we evaluate alternative uses of free cash, but share buybacks have generally been the most efficient use. We began our share repurchase program in 1986 and have purchased Torchmark shares in all years since then, except 1995, the year we acquired American Income Life. In the last 27 years, we have repurchased 74% of the company's outstanding shares.
As mentioned, we expect to generate $355 million to $365 million of free cash in 2013, and if market conditions are favorable, we plan to use most, if not all that cash for share repurchases. Now, I'd like to conclude my remarks on capital management by discussing our thoughts on returning cash to shareholders. For over 25 years, Torchmark management and board of directors have agreed on the importance of distributing the free cash to the shareholders. We have accomplished this through share repurchases and dividends with an obvious emphasis on share repurchases. As a result, Torchmark has consistently distributed a large percentage of earnings to our shareholders. In fact, in the last 10 years, Torchmark distributed 81% of net income to the shareholders.
As for the future, maintaining our cash flow at high levels and returning a substantial percentage of earnings to shareholders will continue to be an important part of our business model. Now I'll ask Larry to make some final comments.
Finally, let's summarize what Torchmark has to offer shareholders and potential investors. First, Torchmark's growth potential. While many people see the life insurance as a mature industry, Torchmark actually operates in a vastly underserved market with little competition, where the majority of individuals have no life insurance or don't have enough life insurance. Our vast experience, along with our ability to control costs, allows us to operate effectively in the middle-income market. With the acquisition of Family Heritage Life, we expect to reverse the recent trend of declining health sales. Our high underwriting margins. Our operating margins are among the highest in the industry. We don't have to rely on investment income to generate profits. Our conservative investment philosophy. Because of our high underwriting margins, we can generate strong profits without having to take significant investment risks. A sustainable mid-double-digit ROE.
Our return on equity, excluding net unrealized gains on our fixed maturities, was 15.5% in 2012. We expect to maintain ROE at around the 15% level. A safe haven in a low interest rate environment. We are much less exposed to low interest rate environment than most of our peers. We are very confident that it poses no threat to our balance sheet. We expect to grow net operating earnings per share close to 10% per year, even in a low interest rate environment. Our insurance operations are relatively immune to the economy. While low interest rates impact our net investment income, our insurance operations have always proven relatively immune to swings in the economy. Even though sales can be more challenging in a difficult economy, our persistency has never been impacted. Our earnings are driven by our in-force block. We have strong, reliable free cash flow.
Our statutory earnings are driven by our large, stable in-force block. Even if we had no sales in a given year, we would still generate over $300 million in free cash flow. Our return of cash to shareholders. As Gary stated, Torchmark has a long history of using free cash flow to repurchase stock. Since 1986, we have repurchased 74% of our stock. We expect that the majority, if not all, of our free cash flow in 2013 will be used to repurchase shares as long as market conditions are favorable. This concludes our comments. We will be happy to answer any questions now.
Okay, I will start with one. You talked a lot about free cash flow and using it for share buyback. This year, also, you were able to put significant capital to work buying Family Heritage. Maybe you could talk about a little bit your view of the M&A environment and what you view as funding sources when opportunities become available.
As far as the M&A environment, the companies that we are interested in, and Family Heritage met all the criteria. Companies that sell in the middle income market, like our other operations do, have a captive or controlled distribution, and have high underwriting margins. They are hard to find. We have been looking pretty hard for the last four or five years, and there are companies that fit that criteria that we are interested in, but they are owned by other companies that are not for sale at the moment. We will continue to look for acquisitions. We like the Family Heritage acquisition. One, we think even though it is small, it has much opportunity to grow. Secondly, we were able to finance that acquisition using, in large part, insurance company funds, which didn't disturb the free cash that we could also use for share repurchase.
When I say the share repurchase has been the most efficient use, like we did with Family Heritage, if we find an acquisition that is going to give us a good yield, then we'll do that as opposed to share repurchase. We're not going to sit on money as the great thing about Torchmark is that free cash comes in every year. When you talk about funding, our leverage is in the low to middle range of the peers, and we have that cash coming in every year. We've got multiple ways to finance a transaction.
When you think about leverage, is there opportunity there if you see something attractive in order to ramp that up a little bit?
Oh, yeah.
Right now our debt cap ratio is around 28%, and once we retire our August 2013 maturity here in August, it will drop down to around 26%. Rating agencies, where we try to really maintain it, is somewhere in that 25%-30% range. We have a little bit of capacity to increase the debt if an opportunity came about.
I'll ask just one more on Liberty National. You mentioned the very strong results this year and the margin expansion. In the presentation, you also mentioned migrating from rural Southeastern towns to more urban areas. Could you talk about what you view as maybe the long-term growth trajectory there, and where margins could eventually hit?
I think the long-term growth strategy is going to grow outside of the Southeast. To first grow, we have to develop our middle management so that we have people we can promote to send to other states. We've started that process. In 2012, we opened an office in the Southwest and the Midwest and the Northeast. What's holding us back right now is just the lack of candidates to fill those positions. It'll be an expansion in urban areas. It's easier to recruit agents in a more urban area than a more rural area. You'll see that steady progress throughout 2013 and 2014 and forward. I think the growth of Liberty National, that's the standard we'd like to see low double-digit growth in our agency force and our sales. That's the standard we'll try and hold ourselves as we go forward.
You're changing the culture of a company, you have to change that culture. We saw a steady progress last year sequentially. I think in 2013 we'll see some year-over-year growth. In 2014, 2015, as we go forward, we'll see that growth expand.
As far as the margin expansion, you shouldn't expect that to pick up another three or four points coming year. It should be around the 26 level, maybe a little bit higher than that. What we did this past year by making the change and moving it more to an independent, although they're still capped, is more of an independent operation where the branch managers are paying responsible for the expenses. That removed expenses that the company was expensing before. That's when we picked up that extra margin.
Next question. When you were talking about the interest rate assumptions used to assess your reserves and sensitivity to continued low interest rates, how did you arrive at that? Is that a mean reversion assumption, or is that based on a roll-forward of your portfolio yield?
As far as we roll forward the portfolio yield, assuming we're earning at 4%, 4.25%. At the same token, though, we've lowered the discount rate on our reserves. Again, I don't want to overemphasize that, but we don't sell interest-sensitive business, so we're not crediting interest to funds or whatever. It's simply we have to discount the reserves. We've also lowered the discount rate on the reserves. As the portfolio yield comes down, also the weighted average discount on the reserves come down as well.
The 4.25% was just a little bit lower than what the actual new money rate that we incurred for 2012. That's why we just went ahead and used that, assuming going forward that should the rates stay stable at the 4.25%, then that, looking at what the effect on the portfolio would be, and then as Gary had indicated, we tested that down at 4% as well.
If rates stay low for a longer period, could you foresee any scenario where you may have to lower the discount rate more and there could be a reserve hit?
No, I don't think there'd be a reserve hit, no. When you lower the discount rate, we continue to have to do that. That can impair, I shouldn't say impair, that it could reduce our underwriting margins. One thing we did, American Income, which is our largest operation, we increased the premium rates on new business going forward by 5% in 2012. We estimated by having to change the discount rate and reserves, we need a 1%-2% increase in premium rates in order to keep our underwriting margins the same. We went ahead and raised them 5%, so we've got a little bit of margin there. The great thing about that is we have the ability to raise the premiums where there's not a lot of competitors. As a matter of fact, American Income really has no direct competitors.
We were able to make that change without explaining to the agents. They had no problem with it. 5% on a $400 annual premium is not a big change. We were able to do that, and that helps keep our underwriting margins where they were. We'll have less investment income, but we're still protecting underwriting margins. We would have the ability to raise, you don't want to do it too often, but we still have the ability to raise premiums in the future.
Just to follow up on the low rates, I know you said for statutory reserves to be adjusted, it would take more than 20 years. Do you have any sense if you ever did have to add to those reserves, how large that increase in reserves would be?
Well, actually, when we're talking about the 20 years, we're talking about GAAP reserves. From a statutory standpoint, the interest rates were very low to begin with to setting those reserves. I can't imagine us having to add anything significant to the reserves. You got to remember, we're talking about 26% profit margins. Even if we had to add to the reserves, it would still be profitable business.
Okay. Thank you very much.