CNO Financial Group, Inc. (CNO) Earnings Call Transcript & Summary

All right. Thank you, Jennifer. Good morning. How's everybody doing this morning? Good. All right. Great. I'm very excited to be here today to talk to you about our product portfolio. So there are 3 key insights that I want you to take away from this presentation. So the first, we have a broad and balanced product portfolio. Second, we have targeted product designs that meet middle-income consumers' needs and enable us to generate profits. And third, we have disciplined and prudent product management. So let's jump in with that first point. We have a broad and balanced product portfolio. In the product space, we're a specialist serving the middle market. We have a well-balanced portfolio of products and a broader portfolio than many of our competitors. We don't just lean in on health or annuities or life. We play across the board to offer the products that are most important to our target markets. This breadth allows us the flexibility to pivot as circumstances change. We not only have these products on the shelf, we're actively marketing and servicing them. This ensures that we retain the muscle memory to pivot if we need to. The insurance margin chart shows you that each of our product lines carries its own weight. Our distribution of insurance margin tracks closely with the distribution of collected premium. Another important benefit that this product portfolio provides is diversification of risk and some natural hedges in the portfolio. The high-low balls on this chart represent the degree of risk present in each of our products. So for some risks, we may have more or less risk present in the portfolio. But for other risks like mortality, there are actually natural offsets in the portfolio. So for example, increases in overall population mortality help us on annuities, and they may hurt us on life insurance and vice versa. So our balanced portfolio provides a competitive advantage through this diversification benefit. And later on, I'll go into more detail on how we manage some of the specific risks on some of these products. So now I want to turn to my second key takeaway. We have targeted product designs that directly address the most critical needs of Middle America and enable us to generate profits. I'm going to touch briefly on our life and health business, and then I'm going to turn to annuities and long-term care. As Gary mentioned, that's where we get most of our questions. In life and health, we offer a standard set of protection-oriented products that are attractive to the middle market. These products are reasonably simple for consumers to understand, and they deliver a solid and stable margins. Note that we don't offer variable life or variable UL. Those are registered products, and they give consumers direct exposure to the equity markets. Now I'm happy to answer any questions you have about our life and health products during the Q&A. But the main takeaway here is that our collective premiums and our insurance margins are stable. So now I'd like to spend a few minutes talking about our annuities. We offer both fixed indexed and fixed interest annuities, although like most other annuity writers, we see most of our sales shifting towards the indexed products. Those products represented 96% of our annuity sales in 2019. So the main difference between fixed indexed and fixed interest annuities is how interest is credited. The premiums that we receive are invested in the same assets backing our annuities, but the fixed indexed product offers consumers access to returns that are indexed to the S&P 500. We declare the participation rate at the beginning of each contract year, but we won't know what the actual performance is until we get to the end of that year, and we see how the S&P 500 performed. The returns are floored at 0. So the customer knows that their account value can never go down. The fixed interest product has a credited rate that's based on our expected earned rate and our general account portfolio, less the spread. The credited rate is also declared each year, but it's known in advance. So the customer will know exactly what they're going to earn on their money. The fixed interest return is also floored at 0 or higher, depending on the specific guarantee in the contract. And I'll point out, also, here that we do not offer variable annuities, those are registered securities. We feel like fixed index is a better fit for our market, as I'll show on the next slide. So our fixed indexed annuities are targeted to our market because they offer a middle path for consumers who want the potential for upside by investing in the market, but without the risk that comes with it. So the graph on this slide shows illustrative performance from a period of July of 2008 to July of 2018. So the blue line shows how $100,000 invested in the S&P 500 in July of 2008, would have performed over that 10-year period. And you can see, it's generally good performance once you get past those early negative years. The gray line shows how the same amount would accumulate at a fixed rate that's similar to what we're offering on our fixed interest annuities today. And remember, that this is based on the rate we earn on our general account assets less the spread. And then finally, the orange line shows how that same amount would accumulate at the participation rate that we're currently offering on our fixed indexed annuities. Now note that that's not the actual participation rate we offered over that historical period. That's why this graph is illustrative. Remember that we reset that participation rate every year, so it would have varied over this historical period. But I think that this graph gives you a good picture of why customers like the returns that we offer on our fixed index product. They don't have to worry about losing value in periods where the index underperforms, as in the early years of this graph. They get the potential for a higher return relative to the fixed return, and their principal is guaranteed. So for our target market, they're willing to give up some of that upside in exchange for the downside protection that we offer. And so by offering what I call this middle way, we not only meet customer needs but we manage our own exposure by passing through some of the equity risks to our customers floored at 0. And I'll talk in a little bit about how we manage that portion of the risk using hedges. Our fixed indexed annuities are also targeted to our market because they are clearer and simpler than the products offered by other writers in the market. We offer simple clear products that they can understand and that we can manage effectively. So I'm showing here a comparison of our guaranteed lifetime income annuity, or GLIA, to another annuity writer's product. And this annuity writer was the second highest seller of annuities, fixed interest -- or fixed indexed annuities in the third quarter of 2019. This is one of the representative products from their suite. We offer our customers a choice between 1 fixed and 1 indexed option. Our indexed option is a simple monthly averaging of the S&P 500, and returns are credited on an annual basis. This other company has 9 indexed options for customers to choose between. And frankly, we see companies out there that offer 20 or more indexed options within a single contract. For customers, choosing between so many options is confusing. And for the companies that offer these, they have to put in place expensive hedging programs to manage such a broad offering of indexes. Second, we don't offer any complicated bonuses. Other companies may offer bonus structures that kick in at different points of the contract. They're hard for customers to understand. They tend to look really good in sales illustrations, but there's no guarantee that they'll actually materialize. Now we offer a participation rate on the S&P 500. Other companies may have a more attractive looking participation rate, but they may also apply caps. What that means is that if the index over-performs beyond what the cap is, they're not going to credit that to the customer. Or they may take a spread, so they haircut the indexed return before applying the participation rate. The point here is that you cannot just look at participation rate when you're looking across this competitive space. We don't charge our customers for our guaranteed lifetime withdrawal benefit, whereas some other annuity writers assess fees for this annually. This is important because what it means for our customers is that their account value can never go down. It's floored at 0 and we don't take charges out. So they have the security of knowing that, that account value will not go down from year-to-year. And then finally, some companies extend the period that customers have to wait before they can take advantage of that GLWB benefit. We allow our customers to access it after one year. So hopefully, I've made it clear how we think our annuities are a better fit for our market. But this market fit isn't just good for our customers, it's good for us. Scott talked earlier about the fact that our Bankers agents sell these products at the kitchen table. They form relationships with their clients. So generally, career agency business tend to have lower voluntary rental rates than business sold through other channels. Also, our agents can only sell our product. So they're not looking to turn customers by moving them to another company. And the customers who tend to buy this product, they're not chasing rate in the market. They're not being courted by the bigger players. They're looking for a product that will give them stability, and so they tend to be stickier with us. So another important point is that our fixed index annuities generate profits. And so let me explain how we make money on annuities. When we receive premium from our customers, we invest it in our fixed asset portfolio. We take what we earn, we subtract out the spread and the remainder becomes what we credit to our contract holders. For fixed interest annuities, we credit that interest directly. For fixed indexed annuities, we use that credited portion to buy options on the S&P 500. And I'll talk more in a minute about how those hedges work. So the spread is our primary source of earnings. It covers our benefits and expenses, and the remainder is our profit. So the charts here show where our spread has been trending recently. The chart on the left is for fixed indexed annuities, and the chart on the right is for fixed interest annuities. The bars show the account value as of 12/31/18 and 12/31/19, and the line show where our margins were as a percent of account value at those dates. So you can see that we're able to maintain healthy spreads on these products even as interest rates have come down. So now let's turn to long-term care. We offer a long-term care product that is targeted to our market. And we offer a product that is less risky in several important ways that differentiate us in the market. The business we're currently selling is predominantly short-term care. We have eliminated or reduced the benefits that drive the greatest risk in this market. So as you can see from the graphs on the right, only 2% of our in-force policies have a lifetime benefit period, and 94% of our in-force policies have a benefit period of 4 years or less. On our 2019 sales, 92% of policies have a benefit period of one year or less and we've eliminated all benefit periods greater than 3 years. We offer a reimbursement benefit, not an indemnity benefit. So we only pay for costs that are actually incurred by our customers. That reimbursement is subject to a daily limit. For most of the policies that we sold in 2019, the maximum limit that our customers could choose was $250. On average, most are at about $150 of daily benefit. Almost all of our contracts are now sold without an inflation benefit. Most other LTC writers are still offering a 5% compound inflation benefit. Long-term care products may have a restoration of benefits feature. What that means is that the full benefit amount is restored, if the customer goes without care for a period of at least 180 days. And benefits can be restored, even if they've already used their maximum benefit period. We offer a onetime restoration of benefits. But some companies are still offering repeated restoration of benefits, some offer unlimited restoration of benefits over the life of contract. So finally, 25% of our sales are reinsured by RGA. And we believe we could reinsure more of this business if we wanted to. But we like the profit patterns on it. So we're comfortable retaining 75% of the business. But what's important about our relationship with RGA is that it does serve to limit some of our exposure. But it also provides affirmation that our underwriting is appropriate, our assumptions are reasonable and the product is profitable. So this less risky product also happens to be a good fit for our target market's needs. You might wonder, if we could still be competitive with a product where we've eliminated some of the richest and riskiest features. But here's the thing, by rightsizing our products and streamlining features, we've actually created a product that better fits the needs of our target market. Our market needs a product they can afford. They can't afford a Cadillac product that will cover every potential future care scenario. So they need to prioritize the short-term care needs that are the biggest concern for them. 50% of all long-term care claims are for a period of less than one year. And the average length of stay in a nursing home for those who are discharged from a nursing home is 9 months. We know that 75% of future retirees don't have long-term care coverage, and that's primarily due to affordability. So by eliminating the richest and the riskiest features from our product offerings, we're able to offer them a product that meets their needs at a price that they can afford. So our success at reaching this market and managing our financial outcomes is clear from our track record. We have stable collected premiums and stable insurance margins, as shown on the chart here. And we have levers that enable us to continue to manage the risk on this block, as I'll talk about in a couple of slides. So the last takeaway that I want to leave you with is that we have disciplined and prudent product management. And so to illustrate this, I'll again talk about our annuities and long-term care. Disciplined management of fixed annuities means active management of credited rates. I spoke earlier about how we set the credited rates. We earn a rate, we take off a spread, and the rest is what we credit. Every month, we review what we're earning in on our portfolio. And we set the credited rate for the coming year for all contracts with an anniversary in that month. So generally speaking, as earned rates move, our credited rates move in tandem. Now all of our annuities offer a guaranteed floor of some kind. The floor is either at or above 0, depending on what's guaranteed in the contract. When credited rates reach the guaranteed rate, we will begin to experience some spread compression. And that means that we can't keep the full amount of the spread that we had planned to. But a lower earned spread does not necessarily mean that we are losing money. It just means we're earning less than we planned to. So we often get asked where we are in relation to these guarantees and how much room we have to maneuver in setting our credited rates. The charts at the bottom of this slide show where our current rates are in relation to guarantees. And while some credited rates are currently at guarantees, particularly on the fixed interest portfolio, we still have room to lower credited rates on 76% of our account value. And again, remember that just because we're at guarantees, it doesn't mean that we're losing money on those blocks. So disciplined management of fixed indexed annuities also requires prudent hedging of the index crediting strategy that we offer to the customer. So as we discussed early on fixed indexed annuities, we don't just pass the credited right to the customer. We use that credited amount to buy hedges on the indexed crediting strategy that we're offering to them. And so the price of those options in the market is what dictates our participation rate. We manage our fixed indexed annuities by fully hedging the index crediting strategy that we offer to the customer. Some companies only partially hedge their fixed indexed annuities. So what they do is they take that credited amount, they use part of it to buy options and they drop the rest of it into profit. That gives them potentially more profits in some periods, but it also creates risk. We are 100% hedged on our liabilities. And as you can see on the graph, that tight matching of assets and liabilities shows that our hedge program is extremely effective. Although the underlying economics are volatile, what matters is that the match remains tight. So next, I'd like to turn back to long-term care. I spoke earlier about how our product designs serve to limit some of our risks, even as they meet our customers' most pressing needs. I'd like to talk about some of the other actions we take to ensure that we have prudent management of this portfolio. So disciplined management of long-term care means knowing when to derisk your portfolio. In 2018, we entered into a reinsurance agreement with Wilton Re that significantly derisked our portfolio by transferring some of our oldest and riskiest blocks. You can see on the charts that we offloaded almost $3 billion of net GAAP liability as measured at 12/31/2017. On the right, you can see that the block we transferred to Wilton Re had a reserve margin of 3% of net GAAP liability at 12/31/2017. You can also see that the block we retained, which currently has a net GAAP liability of $2.4 billion as of 12/31/19 is sustainable and healthy in total. Our reserve margin on this block has grown from $234 million at 12/31/17 to $255 million at 12/31/19 in spite of interest rate headwinds. All right. So this is where management of long-term care also means prudent reserving and active management of operations. So the first point I'll make here is that our reserves are calculated using prudent assumptions informed by historical experience. We don't assume morbidity improvement. Other carriers sometimes do, and it can be overly optimistic. We include some portion of rate increases that we filed for but that we haven't yet received. Other companies may include rate increases that they haven't even filed for and that they may not have actuarial justification for. Also in our 2019 testing, we set what we believe is a prudent ultimate new money rate, which was at a 4% 10-year treasury. Second, the economics of our block growth is strong. As I showed on the prior page, although we use prudent assumptions, our 2019 loss recognition testing margin increased to $255 million or 11% of net GAAP liabilities. We have never taken loss recognition testing charges on our open block. Third, we have actively managed the experience on this block by taking rate increases. We've taken rate increases, when warranted, over the 20-plus years that we've been managing this business. Over the last 10 years, we've had 2 rounds of rate increases in 2010 and 2015. On average, we ask for 20% rate increases and we receive 10%. Finally, we actively monitor our operational processes around long-term care. Our reinsurance agreement with RGA ensures that we maintain appropriate underwriting discipline. We leverage outside key experts, like LTCG, to help us manage our claims and help us understand best practices in that area. And finally, we actively monitor our internal processes around claims, fraud and terminations. So as we look ahead to 2020, we'll continue to maintain these same themes in the management of our portfolio. With the shift from 3 segments to 2 divisions, we'll be looking to maintain balanced product portfolios in the Consumer and Worksite divisions. We're continuing to develop and launch new targeted product designs that are good for our customers and enable us to generate profits. And we'll continue to maintain our disciplined and prudent management as we evolve our accounting and reporting in response to the GAAP requirements under long-duration targeted improvements. And so with that, I'd like to turn things over to our Chief Investment Officer, Eric Johnson.

All right. Good morning, everybody. And why don't we just jump right into it. We have a lot of information to share with each other. Like everyone else, I get 3 key takeaways. Key -- number one, we're managing through the low-rate environment, I think, very adequately. One thing we're not doing is compromising on what counts for the long term. And what that means is that in the future, risk level we have in the portfolio may go up, and it may go down. But it will only go up when we're being adequately compensated for that, and it's good for the company, we have the capital to support it and it's good for the shareholder. I don't think that condition exists today. Second key point, our general account is in very solid shape, and it turns out quite predictable income. Third key takeaway, I think we're extremely well positioned for the uncertainty of the times and the turn of the cycle. Stepping back from the 3 key takeaways, in the big picture of things, when everyone leaves here today, I really don't want you to talk so much about interest rates and spreads and things like that. Really focus on all the good things Mike and Scott and Gary are doing to really build the sales machine of the future. That's the real story of today. But what I want you to go away with today is confidence that we've got the money in the right place to make the future happen for the company. Talking about rates. I remember when I first started working, the 10s were between 14% and 15%, yield to maturity. Prime rate was between 18% and 19%. Who even says prime rate anymore? So times have changed quite a lot. But one thing that hasn't changed is, if you're in the investment business, you learn how to fly the plane through all kinds of different market cycles and conditions and land it successfully. And we're going to do that here also. We have a very methodical process with a lot of folks with gray hair or at least hair. But one of the things that's cool about it is, it's largely about -- the machine we've got is, in many ways, largely about managing how yield and risk and credit interacts with the rest of the enterprise. We have a lot of resources and spend a lot of time and energy on risk planning and capital planning, strategic asset allocation and certainly, asset liability management, as Karen was just describing. And if you looked at the tenure of those folks on our team who are involved in all of the exercises, I'm the oldest one, probably, but not by that much actually. Enough about me. Let's talk about the company. And certainly, the topic of the day is interest rates. And there's one inescapable, which is that as long as new money rates are below book yield, both book yield and net interest income will be pressured. That's the force of gravity. Let me give you some background on kind of the slope of the book yield line, which you see in front of you, which is the top line. And over the last couple of years, it's down roughly 35 to 40 cumulative basis points. Let me -- I can decompose that for you a little bit. Of that 35 to 40 basis points, 15 of them were the long-term care transaction that we did to reduce risk in the long-term care block. That's the largest factor. About 10 of them were a trade we did last year, which we call up-in-quality, which -- more on that later. And the rest of them, roughly 20 are just the run rate drip of low for longer, 3 or 4 basis points a quarter. So a way of thinking about this would be the following. If we turn our portfolio over actively at about 1% per quarter, and we get in roughly 2% of the portfolio per quarter in maturities, prepayments, repayments. That is up to around 3% of the portfolio coming in per quarter. And there's net new sales from all the things that Scott and Mike were describing. So it's 3% or 4% per quarter of the portfolio is coming in. If you take the delta between the new money rate and book yield, let's say, it's a 100 basis points, a 100 basis points times 3.5 is 3.5 basis points. That's roughly the delta in book yield quarter-over-quarter. It's actually a very simple formula. It's just gravity. From a managerial perspective, keeping our voluntary turnover down to around 1% a quarter or 2% a quarter. Imagine a large room full of hyperactive investment professionals like yourselves, all sitting on their hands all day. That's basically the management task. And we've had turnover as much as 40% of our portfolio in a year when market conditions justified it. We've had it as low as 1% a quarter, last quarter. So we try to do what's good for the company, not what's fun for us. We manage according to the circumstances. You can see on this chart, it's on your left-hand side, the onetime impacts on book yield from a long-term care transaction. You see the 15 basis points. And from what we're calling the up-in-quality trade which is, you see there the 9 basis points. But what the bigger picture of this page is that basically book yield times AUM creates core income and it trends very predictably. One question you might ask because certainly it's one question I asked myself is the up-in-quality trade, would I do that again? The answer is yes. And I'll give you the reason. If we're foregoing roughly $10 million to $15 million in income a year, I view that in the nature of an insurance premium. And what we do know is that ultimately, the house will be on fire. And what I can tell you is that I think when that day comes, that $10 million or $15 million a year in foregone income will be viewed as a very, very -- the present value of that transaction, I think, will be very successful. It's good today, for example, to be basically half-weight energy, to be underweight airlines, underweight containers, underweight railroads, underweight auto, underweight metals and mining. More on that later. Let's talk about variance. We have 2 sources, principal sources of variable income in our general account. The first one is, spoken broadly, from alternatives, which are nontraditional asset classes. Yet we have roughly $580 million AUM in that area, and I'll say some more about it later. If you look at the rolling 4-quarter average quarterly income there. It revolves around the low teens. But it has variance. It can be -- it's been as low as, over the last couple years, $5 million in a quarter. It's been as high as $18 million in a quarter. But the average is pretty clear. By the way, when we're planning for income for a year, we tend to use something around 8%. So it's pretty much always done better than what we've intended it to do. Although it does have some quarterly variance. The second area where we generate variable income is prepayments and calls. Now this is obviously involuntary. And what we do is receive it. We can't -- we don't manage it. Having said that, a good rule of thumb to think about would be 2 to 3 basis points per quarter. And that's an average. As you can see, you've got a big bar that says 10, a little bar that says 2. And the reality is if you stand too close to it, it's hard to see. But if you step back, you can see the normalization. And that's how we think about it when we plan it, and that might be a way for you to think about it when you model it. One of the key messages I want you to take away today is that our core income, which is book value times AUM is consistently over 90% of our NII. It's very consistent. Secondly, that the noise in variable income, which I just described, has a much lower effect on NII than perhaps you think. 50% changes in variable income creates about a 3% change in NII. So don't focus on the 50% focus on the 3%, right? And that's a very -- it's quite transitory noise, although I do recognize that in terms of its effect on quarterly EPS, it leverages its way down the balance sheet. If you look at the area I boxed there in red, which was the second quarter of 2019. That had a sort of -- I'm calling that the harmonic convergence. It's was all good, pretty much everything worked the way you would ideally like it to. The up-in-quality trade was just getting done. So we were still earning all that income. We had a good quarter in alts, and prepays were a little bit over their norm as well. So it was a really good quarter. There'll be another one of those out ahead of us again, but it's not going to be like that every quarter. So if you're wondering why did income step down the second parts of the year, harmonic convergence doesn't last forever. You appreciate it and look forward to the next one. We have a quite a traditional portfolio diversified by sector. We have very consistent key risk measures, which we track, probably about 25 of them on a monthly if not more frequent basis. And we're actually pretty consistent with our sector allocations. Although there is much more activity within the sectors than there is across the sectors. And what this tells you is that we're willing to be consistent, and we're willing to be boring as long as we can also be successful in generating the core income the company wants. And I think we're doing all those things. Another area I get a fair number of questions about is in the structured securities area, particularly around the ABS space. This is roughly 1/4 of our portfolio, which is a little higher than many companies. 97% investment grade, it's a really good fit. Karen described the annuities, and how we've had good margins in that area, in part because we invest heavily in this space. Their credit duration matches up well with fixed income annuities. They have good non-binary outcomes. The market volatility is a lot less than many other fixed income spaces because they're shorter on the curve. They tend to have heavier coupons. They tend to have amortizing structures or prepayable structures, so you're getting cash back from day 1. I'll say, talk a little bit more about it later, I think, but it's a diverse mix of asset-backed securities, RMBS, CMBS and CLOs and some other specialized areas as well. We like it very much. We have a lot of skills in this area. In fact, I'd like to invest more in it, if the opportunities present themselves. We have a commercial mortgage portfolio, which -- this is an area where we like to be boring. We have very conservative underwriting standards. For example, we don't do any construction lending. We don't do any non-stabilized properties. We don't do any raw real estate lending. And we hit it right down the middle of the fairway. CLOs is an area people ask me about often, and here's what I will tell you, we operate with great caution in this area with a true eye to the downside. Our debt portfolio is A and better only, with a strong margin for safety. If you believe Moody's and S&P, which I think you can, no A or better CLO tranche has ever lost money. I don't think we'll be the first to test that maxim. Our equity portfolio, we think, is sized to our tolerance for risk. It's managed. We manage that, those CLOs. We tend to concede on the equity distributions to make sure what we have is high-quality collateral. This is an allocation we've reduced 40% over the last 3 years as the ARB has decreased. We'll tend to put them on when the ARB is producing mid-teens or higher returns. Down here where it is at 8%, 9%, 10% doesn't pay, so we don't do it. So we've actively managed this according to the company's need for return and eye on risk. And I'm very pleased with our results in this area. I mentioned earlier in alternatives that we have roughly $580 million allocation. I used the term, I'd like to focus on strategies that pay rent. And what I mean by pay rent is, they're just making distributions. There's no J curve. They're making distributions from day 1. Core real estate, infrastructure and private credit are 3 good examples. And those are areas where we have large allocations, and it's produced fairly consistent, stable returns for the nature of this asset class. It's produced less in -- on the upside than stocks. But it has not experienced kind of downside velocity we've seen in stocks the last 2 days, and it pays a heavy current. As I described earlier, we model-in 8% as a run rate. And I think the current income on this portfolio, the carry is north of 8%. So I think we have a good reasonable view on it. As I said earlier, we are pretty consistent in our attitude toward risk. And we have, over long periods of time, pretty stable key risk metrics, which we monitor very closely. We consider this in close linkage with enterprise capital strategy. And we try to be conscious of competing needs for capital or opportunities for capital around the organization. People often ask me, how do we think about possible allocation changes. We think about it all the time, but we think about it in this context. We don't think about it in isolation from the rest of the needs of the company. As Karen described to you earlier, one reason we've had good annuity results and done better in long-term care than people think we have a right to is, we have a very disciplined approach to asset liability management. I will tell you, we manage at a far higher degree of granularity than we report. For example, you think we have one general account, or you think we have 3 statutory entities. But from where I sit, we have roughly 17 portfolios, which are product portfolios within those entities, which we managed according to the needs of each of those independently. Now it all has to roll up into something that makes sense, but it has to make sense at the granular level as well. And so let me say a word about low rates. Low rates have exerted a price, but it also -- that environment also has its benefits. The housing market benefits greatly. Commercial mortgage market benefits greatly. Custody credit benefits greatly. Emerging markets benefit greatly, and consumer credit generally benefits. All of those benefits mean there's less credit dilution. Less credit dilution helps offset lower NII. So -- and putting that all aside, we're sitting on close to $3 billion positive AOCI today. All of that is a benefit of having low interest rates. Having said that, NII helps feed the beast, and we like it very much, but I do want to mention that there are benefits to the climate we're in. My last page is really around kind the macro theme, but it's hard to talk -- which is about quality assets and stable income. Hard to talk about anything, though, in the year 2020 without mentioning recent events in coronavirus, and how are we thinking about that? And it's a human tragedy of unknown dimension. And first, we're an insurance company. We want people to be well, stay well and get well, and we get paid to help them along that journey. And that certainly comes up ahead of investing money for sure. But having said that, most of the sectors that are most greatly exposed or so they say to the current circumstances are sectors that we're -- fortunately, we've been pretty light. We have a very, very small exposure to Asia. Asia and EM directly probably under $50 million. That includes China. Now someone's going to stand up and say, yes, but Chinese fixed income market has done great the last month because of what the government has been doing. True. We're pretty light commodities. We probably have about $120 million in metals and mining in total in the big diversified global names, which will be fine. In terms of autos, all we have is GM, and it's a small amount of that. Containers, we are not a huge investor in container securitizations. We only have less than $50 million. And in energy, as I mentioned earlier, we're probably about half weight. So whether it's by brains or luck, I think we're fairly well positioned for the circumstance we have to the extent we can understand the circumstance we have. I mean, there's a lot yet to be experienced. And the effect on the global economy and just on the local circumstances, we'll just have to wait and see. So that's my prepared remarks. And then on that note, I'll turn it over to our cleanup hitter, which is our CFO, Paul McDonough.

Thank you, Eric. Good morning, everyone. I don't have a joke to start with. I'm sure those we manage you who know me expected me to start with a joke, but I don't have one. Thank you so much for being here. We really appreciate your interest and in most cases, your support of CNO over the years. We are, as I hope has been evident in all of our prepared remarks, we're proud of what we've accomplished in 2019. We are excited about the opportunity that we have to continue to serve our target market and, we think, create value for our shareholders. The focus of my prepared remarks are to provide hopefully some perspective, some insight into our 2019 financials and also provide some outlook, a bit more than we have provided in the past, maybe not quite as much as Gary alluded to in his earlier remarks. And in the process of proving that perspective and outlook, I'm hopeful that I'll leave you with 3 key takeaways that are summarized on this page. So the first point is, as we have established what we think are sustainable top line sales, we're now turning our attention to capturing operating efficiencies: i.e., actually reducing expenses year-over-year, which will offset the headwinds that Eric was just talking about from lower interest rates and will result in a better balance between sales growth and earnings growth. The second point is that continued strong free cash flow and disciplined capital management will support the 9% to 11% earnings per share growth that Gary mentioned in his opening comments. And the third point is that our disciplined stewardship of the balance sheet will continue to optimize our assets and ultimately enhance shareholder value. As I think most of you know, we are introducing today a new reporting framework. We posted on our Investor Relations website this morning, a new investor supplement that recharacterizes our '18 and '19 financials in this framework that's summarized in this chart. We also filed that new supplement in an 8-K, and with our first quarter 2020 reporting we'll be formally adopting this with our SEC filings. So let me just walk you through this chart. These are the main components of operating earnings under the new framework. The first component is insurance product margin by major product lines, so life, health, annuities, long-term care. The second component is expenses that we allocate to insurance products in total, but that we don't allocate to the individual product lines. The third component is fee income; fourth, investment income not allocated to product lines, so not embedded in insurance product margin. And then lastly, expenses not allocated to product lines, largely truly corporate overhead. The sum of all those things is operating earnings. We do believe, and we hope that you'll find that this new framework will provide more transparency and more insight into the profitability of each of our products. You'll see that the -- each product contributes meaningfully to our earnings. I like the fact that it distinguishes -- this new framework distinguishes between different types of expenses, variable and fixed and corporate overhead; and different types of income, net investment income, that related to the assets, the fees, the insurance liabilities and the rest largely relating to our capital. So in addition to that, look at operating earnings. The new supplement in the new reporting will show top line sales growth for each of our products grouped by the Consumer Division and the Worksite Division. And as you can see here, and just echoing Scott's remarks and Mike's remarks, we've clearly established top line growth, which, again, which we think is both sustainable and profitable. All right. So in the next few slides, I'm going to provide a little bit more information about each of the major components of operating earnings in this new framework. The first being insurance product margin, which you can see in the chart on the right, grew by 2.8% in '19 versus '18, reaching roughly $890 million. The components of insurance product margin, as defined are: policy income; net investment income allocated to product margin net of insurance policy benefits, amortization and nondeferred commissions; other nondeferred expenses such as advertising, particularly supporting our direct-to-consumer channel; and where applicable interest credited on our annuities. The way that we allocate net investment income to each product line is using the book yield of the actual investments that back the block of business applied to the average insurance liabilities for the block in each period. Notably, insurance product margin as defined here is not reduced by the expenses that we allocate to products in total, but not to the individual product lines. That's all I have to say about that. The next item is fee income. As you can see here, as Gary and Scott alluded to, the income has grown at a very healthy clip from '19 to '18. I should point out that in the fourth quarter of '19 due to a change of assumptions in estimating lifetime revenue for revenues and expenses related to Med advantage, the fourth quarter '19 fee income was increased by about $6.5 million. But even net of that, a very healthy growth. As I think Scott and Gary alluded to in the Q&A, we like this source of income. Certainly, from an ROE perspective, but also from a tax perspective, as it's a source of non-Life income relative to our non-Life NOLs, which I'll touch on in a bit more detail on a later slide. All right. This is the largest bucket, if you will, of expenses. So these are expenses that are not allocated to each of the product lines. They're largely fixed in nature. These are the -- it's the infrastructure supporting our businesses and it's the shared services that also support our businesses. And I would point out sort of 2 dynamics that are operating here that are captured in the blue boxes at the top. One is the fact that expenses are increasing, albeit at a decelerating rate, if you look at the chart on the left, which breaks '19 versus '18 into first half and second half. The reason that they've been increasing is that we have certain expenses that I would characterize as mandatory. So think of things like the cost associated with preparing to comply with long duration targeted improvements which the industry will adopt in Q1 of '22. It's extraordinarily expensive to do what you need to do to be prepared to adopt that new reporting. We feel good about where we are. We feel like we're ahead of the curve in being prepared, but we're spending a fair amount of money as is all of our peers. Second example of mandatory expenses are expenses associated with complying with the California Consumer Privacy Act, also very expensive in terms of compliance. The third example is everything that we're doing and the industry is doing generally to help manage cybersecurity threats. We spent close to $20 million in 2019 on just those 3 things alone as compared to less than $10 million in 2018. So that's certainly -- those kinds of things are certainly contributing to the increase in expenses. The other thing that's adding expense are all the decisions that we've made to invest in growth. Scott spoke to this. Mike spoke to this. And we certainly think that those investments are paying for themselves, helping establish sustainable top line growth, which importantly is also profitable. Having said that, as we've mentioned a few times this morning, we're now at a place where having established sustainable top line growth, we're turning our attention to reducing expenses. And you've seen a couple of examples of that in the last few months. In November, we issued a press release announcing a strategic technology partnership which will reduce expenses by about $20 million over the next 5 years. In January, we announced our new operating model, which we do believe will be truly transformative. We didn't adopt that model for expense reasons. We adopted it to adapt to what we think are changing consumer behaviors and expectations. But one of the things that we expect will happen, we know will happen, we've implemented this, really, already, is when we go from 3 segments to 2 divisions and simplify our structure, we take expense out. In total, we're taking out about $22 million of expense from our cost structure. We're reinvesting about half of that in the business. So a net savings of $11 million. Beyond those 2 things that you've already seen, as we went through the 2020 planning process, we simply just tightened our belts. As we're going through the process and going around the table with the executive leadership group, everyone had to contribute. Beyond that, we've, I think, been thoughtful about identifying specific initiatives that would take cost out. And so we're actually expecting that expenses in '20 versus '19 will actually go down. These are -- this is the last bucket of expenses. This is really -- it's really corporate overhead, not allocated not embedded in product margin, not allocated to products in total, things like public company expenses, board related costs, some portion of executive comp, that sort of thing, trending down, reflecting our expense management discipline. All right, turning to investments. So in this new framework, there are 2 buckets of investment income. The bucket on the left is in investment income that's allocated to the insurance product margin. We've deliberately kept the scale on all 3 of these charts the same, which highlights the fact that the bucket in the middle, which is the net investment income not allocated to product, is tiny compared to the bucket on the left. The assets supporting net investment income allocated to insurance product margin are primarily fixed income with some -- a portion of our alternative investments, largely allocated to some of our longer duration products. The balance in the middle is mostly what Eric described as the variable components of income. So the lion's share of alts. Interestingly, and you could not have seen this in our old reporting, the net investment income allocated to insurance product margin actually increased year-over-year by about by about $14 million. And there are 2 dynamics happening there. One is the insurance liabilities actually increased by about $0.75 billion in '19 versus '18. And at the book yield on the portfolio in '19, that drove about $38 million of incremental NII. Going the other way is the fact that book yield is declining based on the gravity that Eric described. And so the net investment income on the beginning portfolio was lower than what it produced in '19. And that went the other way, the net of those 2 things was the positive $14 million. On the other hand, the bucket in the middle went down by about $42 million, which is largely a reflection of the volatility that you're going to get in that space. And so that's ultimately the gravity that Eric was describing. So putting that all together, I spoke to each one of the major components of operating earnings. You can see that in '19 versus '18, we had growth in insurance product margin, growth in fees, and that was offset by higher expenses and the headwinds from net investment income. If I think about what's our conclusion about that. As I mentioned at the outset, we're actually very proud of what we accomplished in '19. And notwithstanding this total dynamic, if you just think about the macro context, both in terms of lower interest rates and also in terms of where we have been in our journey, focusing very much on establishing sustainable top line growth and sort of during '19, pivoting to a more balanced sales and earnings picture. Okay. Turning away from earnings and focusing on capital. I think as most of you know, we've been reevaluating our target capital, largely in the context of the very different risk profile of our business in the wake of the long-term care reinsurance transaction that Karen talked about. My thesis, when I joined the company in the second quarter of last year, and Gary was certainly pushing me in this direction, given the reduced profile that we ought to at least look at our target capital. And you would think that in the context of a lower risk profile that the target capital would go down. But we didn't want to make that kind of a decision without really being thoughtful and doing some analysis. So we've done that analysis and have concluded and have decided and are sharing with you today that our target capital inside the operating companies, we're reducing from what it had been, 400% to 425%, down to a new range of 375% to 400%. All else equal, that frees up about $125 million of capital. We, in arriving at that decision, certainly worked through the rating agency capital models to make sure that we, and each of the rating agencies were aligned on those models. There is some redundancy in each of those models relative to our current rating. We also used our own internal capital models to have an independent view of what would be appropriate in terms of target capital, largely shocking the balance sheet with stress tests and having an informed view of how much of a cushion we should have. And lastly, we looked at peer companies and their publicly stated target capital. And from the standpoint of all of those things, concluded that this new target was appropriate. At the holding company level, we're not changing our previous target, which was minimum holding company liquidity of $150 million. That's about 2 years of interest expense and about a year of parent company expense. That seems an appropriate level. We're not changing that. We're continuing to target leverage, so debt to total capital of 22% to 25%. We're comfortably in that range and will remain so. I think as all of you will recall, we refinanced a portion of our debt during '19. And so our debt capital is, we think, in a good position in the wake of that refinancing. And then lastly, we remain very committed to our investment-grade ratings. The company fought long and hard to get back to investment grade. We were upgraded by Moody's on the heels of the long-term care transaction in '18 and and by S&P and Fitch in the middle of '19, but largely also in reaction to that transaction. I should say that when we were reaching the decision to reduce the RBC target, we did talk to each of the rating agencies to try to make sure that we were aligned with them on that. We don't have a bright line answer from them because that's not how they operate. Bob Garofalo and [ Erica Reynolds ] are actually here. And so they could tell you themselves if we ask them that we don't have a bright line. But in the context of the discussion we have, in the context of redundancy in the models, in the context of what we think will be continued stable operating results, I'm very confident that this reduced capital doesn't have any adverse rating implications. Free cash flow. So as mentioned on the previous earnings calls. For the full year 2019, we generated about $330 million of free cash flow on a gross basis and about $290 million at the holding company, net of the capital that was retained in the operating companies to support our organic growth. Relative to our operating income of $290 million that translates to free cash flow conversion of close to 100%, which I think is probably one of the more compelling aspects of our financial profile. We will continue to be disciplined and opportunistic in our approach to capital allocation. Those of you who tuned into our earnings calls will be very familiar with the left side of this slide. Captured on the right side, in 2019, as Gary mentioned, we returned $319 million in capital to shareholders through a combination of ordinary quarterly dividends and a share repurchase. That's 13% of the market cap at the beginning of 2019. As Gary mentioned, the highest level of capital return since 2015. And over the last 3 years, we've returned over $700 million of capital to shareholders or 25% of our current market cap. In 2019, we also closed on the Web Benefit Design transaction that Mike talked a lot about, which certainly improved our market presence and capabilities in the worksite space. As stewards of the balance sheet, we're always looking for ways to maximize the value of our assets. And as a great example of that: in 2019, no news, but I just wanted to emphasize the tax strategy that we adopted in the fourth quarter, which resulted in a reversal of the tax valuation allowance that we had to the tune of $194 million; which flowed through net investment income in the fourth quarter and increased our book value per share at year-end '19 by $1.29; contributing to a 13% growth in book value per share ex AOCI for the year. As of year-end '19, we had about $0.5 billion of life NOLs and about $2 billion of non-life NOLs with the combined economic value of just over $400 million using a discount rate of 10%, which equates to $2.71 per share, which I'm nearly certain is not fully reflected in our current stock price. Also, as stewards of the balance sheet, we're, I think, appropriately conservative in setting assumptions and recording our liabilities. I think this is evident in the very modest impact on earnings in 2019 from the annual review of actuarial assumptions. Despite lower initial earned rates and reduced ultimate new money rates, we had no loss recognition triggered in our FAS 60 products, reflecting, I think, very healthy reserve margins in those products; and a modest $10 million of unlocking in our FAS 97 products. I'll conclude now with our 2020 outlook, noting again that we've elected to provide some more guidance than we have historically, perhaps not as much as some would like. As we've said a number of times now, we expect earnings per share off of $1.80 base; significant items in 2019 to grow by 9% to 11% in 2020, obviously, benefiting from share repurchase. We expect earnings in absolute dollars to be flat to slightly up, with continued growth in sales, insurance product margin and fee income; a decrease in expenses, as I mentioned earlier; offset by continued headwinds from lower interest rates. Based on net free cash flow at the holding company between $330 million and $380 million, we expect to have the capacity, not necessarily the commitment in the context of how we think about capital allocation, but the capacity to deploy $75 million in share repurchase per quarter, absent compelling alternatives. So that's it. Thank you very much for your attention. We look forward to another Q&A session. But first I'll ask Gary to come back to the stage.

Thank you, Paul. All right. I am the final impediment between all of you and lunch, I guess, me and Q&A. So as I wrap this up, I want to begin by just, again, placing in context where we've been. If you look at the company and just step back a moment to think about some of the things we've done. The first thing that I want to ask you to remember, and I've said this a number of times on our earnings calls, please judge us by our actions. I think we have a story to tell, and I think we have a certain amount of credibility where we can say, we say what we do and then we do what we say. We try and keep it very simple. And if you look at what the company has been through prior to 2017 when we were really focused on the turnaround and de-risking the balance sheet and cleaning up various things, there were a number of things there that we did that I think were absolutely critical and really positioned us for the future. And then in 2017, roughly, we started to pivot towards growth. Now there were a couple of other things there that we did such as the LTC transaction during that period. But largely, we were really focused on getting that growth engine going again. And I think we've demonstrated that. Now we've had several quarters of consistent top line growth so we feel like now that we've built the growth momentum back -- we cleaned up the company. We've built the growth momentum. That's up and running, now is the opportunity for us to maximize profitability. And again, I think one might be tempted to say, well, why couldn't we do it all at the same time. And I go back to where I started today's meeting. Based on my understanding of how insurance works and having been in this business for 30 years, it's not possible to take -- to cut expense, grow sales, have a net investment income headwind and still drive earnings. You can't do all 4 of those things. There's a certain amount of sequencing that just has to happen, particularly when you're facing net investment headwinds like we've been facing. And so I think that the sequencing was the right way for us to do this. And so now we turn to 2020. We've been pretty clear about some of the expense efforts. We signaled those. We announced them in late '19 and early 2020. We're now executing against them. We've been pretty clear about our capital and how we view the less risky nature of the company. And Paul, I still disagree with you. I think the amount of guidance we're providing is a ton. At least a ton more than I'd like to. But that's a different discussion. I think we've had a demonstrated ability and willingness to be thoughtful about how we deploy that capital. And I think we've done that in a shareholder-friendly way. And so now we look at 2020, we've got a number of different opportunities. You heard from the businesses in terms of how we're bringing some of these assets together and what that represents. So that's all fine and good, but what does it mean? You all have a lot of things you can decide to invest to -- in. And some of those choices may be within the insurance sector, some maybe outside the insurance sector. But at least, as you think about your options to invest in the insurance sector, I'd ask you to think about a couple of things when you're looking at us against other enterprises. First, about 90% of our earnings come from controlled distribution. Any company will tell you that the more business and more earnings you get from that controlled relationship, the stickier they are going to be. There's no surprise there. You don't have the same pressure to churn. That agent is tied to that one manufacturer. And 90% of our earnings come that way. I think the other thing that we've talked a little bit about, but I want to be really explicit on, when we think about that notion of controlling the client relationship, remember that when you sell an insurance product, whether it's auto or homeowners or Life or LTC or whatever it is, for every consumer, that insurance product is typically an expense. And what do we want to do in life with expenses? We want to minimize them. When you create a relationship with a consumer where you are getting assets from them, whether it's in the form of an annuity or they're investing -- the parking assets at your broker-dealer. When you have assets, the relationship fundamentally changes. You are no longer an expense to be minimized, you are an asset to be grown. And I always come back to this point. There's a fundamental change that happens because even if that consumer decides they're going to move away from you, if you've got one of their annuities or they've got money in a mutual fund with your broker-dealer, at the very least, you're going to get a chance to get that last phone call to try and hold that relationship. You don't always get that when you're in insurance sales. 2/3 of my career has been spent as an agent, and I know this the hard way. Sometimes clients just move. But if you have something that's an asset, the nature of the relationship changes. So I really want to emphasize this point, 90% of our earnings come from controlled distribution, and even within that controlled distribution, because the product mix is changing, and we're doing more things like annuities and broker-dealer, even within that, we're making the relationship that much more likely to be sticky or persistent. 65% of our business is coming from life and health protection products. Now in the last several months, part of the reason we've made this change with the reporting, it became evident to Paul and me that through these discussions, there really wasn't a lot of clarity on where our margin was coming from or an understanding of our product margin. I think there's a much more robust mix here than maybe is appreciated it first. So that's part of the reason we adopted that change in reporting. And I think when you look at how much of it, 2/3 coming from life and health, there's a lot of volatility that's frankly isolated by virtue of this mix. This is a good thing. 99% free cash flow conversion in 2019. We continue to have an ability to generate strong cash flow, and we'll continue to be disciplined and opportunistic in how we deploy that. As important as the agent relationships are, and as much emphasis as we place on that, we're also mindful of the fact that consumer behaviors are changing, that they expect to have research or purchase or other decisions made online or on the phone and so on. And we're adapting to that. We've got a heck of a machine at Colonial Penn in terms of using analytics, in terms of understanding how to interface with the consumer on a direct-to-consumer basis. And with our transformation to that Consumer Division, we have an opportunity to leverage that expertise throughout the organization and really share leads and drive things between all the different customer touch points. And then finally, our Worksite business. Mike talked about the individual assets that we have there and the opportunities that we have there. Worksite is one of the fastest-growing distribution models within the insurance space. And you can see that in the way certain companies are valued in terms of the multiple they attract and so on. We ourselves have had 6 consecutive quarters of double-digit growth. Now granted, it's been a relatively small base, but we're ramping that up, and we're extremely excited because with the addition of the technology platform and the former Web Benefits Design, we really feel like we've positioned ourselves to be one of the few players in this space that, like our Consumer business, we're focusing on a smaller client, middle market consumer. And in this space, we have the manufacturing, we have the distribution and now we have the technology. I think we're extremely uniquely positioned. So as I wrap us up, maybe just one quick thought that's not on the slide, just to kind of summarize. If you're trying to decide whether or not to invest in this company and be a part of the story, I'd ask you to think about just a couple of really simple things. I've talked about the positioning and the key takeaways and the cash flow and all that other stuff. But let's make it even simpler. This is a company that's operating in a space where there's a tremendous secular tailwind. Meaning, you've all seen the stats, 10,000 people retiring every day, the whole baby boomer demographic, all this sort of stuff, right, we're serving a market that will continue to grow for some period of time. And within this growing market, we've got a niche called the middle market. Everybody talks about serving the middle market, but no one is focused the way we are. And it's hard to focus on this because even if you've got the cost structure to do it, the reality is, if you've got salespeople out there that like calling on $2 million clients, they don't want to go focus on the $100,000 client. And that's just simple human nature. So we've got this secular tailwind. We've got a really unique focus that nobody else is really getting after the way we are in terms of just being focused on that middle market, and we're using that focus or we're getting into those relationships, and we're staying in through a unique combination of being able to do it face-to-face or direct-to-consumer online. While we're doing that, we're throwing off a lot of cash. And the beauty of that cash is even when we enter economic times where our net investment income is challenged, and therefore, the ability to generate higher absolute earnings dollars is challenged, we can still drive EPS because of the amount of cash we're throwing off. And so as you think about some of the things we've talked about that are unique to us, and if you really boil it down that simple and think about the secular tailwind we have, think about the niche market focus, think about our ability to get in and stay in, think about our ability for the cash and then how we deploy that cash. I don't think you're going to find a lot of folks that are going to tick all those same boxes and have all those same assets. I want to thank you again for your support and interest in CNO. I think we're going to move to a Q&A time. So with that, I'll invite my 3 colleagues back up, and we'll go from there. Thank you.

Humphrey Lee from Dowling and Partners. Gary, I appreciate your comment about the -- how you anchor your customer relationship with the benefit of annuities and mutual funds. Can you maybe kind of give us some color in terms of how much of your existing customer base is also owning a -- whether it's a fixed annuities or mutual funds products through your broker and dealer network?

We've had our annuity balances go to $9 billion right now. And our -- the assets at our broker-dealer is $1.5 billion. So as a measure of total assets, that gives you a picture. In terms of how many of the households that we serve, I don't know if we've got that data handy, I would guess it's, I don't know, less than 20% for sure in terms of have -- using our broker-dealer. Okay, single-digit percentages on the broker-dealer? And of the households we serve, what would we guess is the penetration in terms of how many of the households we serve have annuities with us right now? Single-digit percentages. So call both of those sub-5% in terms of what the current penetration is. Now the way I look at that, and I tend to look at the glass half full. I think that represents the magnitude of the opportunity we have, that the vast majority of households where we have a relationship where they trust us, we haven't yet penetrated with our broker-dealer and/or annuity offerings.

So I guess, my follow-up question is, what action plans do you have in order to capture more of the penetration? For like -- is it simply by having more agents that is licensed to sell securities? Or is there something else that you need to do in order to get that penetration?

Yes. There's a couple of things I want to mention about that. So first of all, as we've demonstrated, I think, through multiple quarters, if you think about our actions at Colonial Penn, if you think about our annuity sales results last quarter and so on, we are perfectly fine taking our foot off the gas pedal on a given product set or a given customer set, if it doesn't make economic sense to us. Now in the case of annuities, I tend to think we'll do okay this year. I don't think the growth will be nearly what you saw last year. That's in part due to the fact that we're coming off of such a strong base from last year and in part because of market conditions. But what we're going to do with annuities, we continue to remain committed to the product. We think it's good. We think our customers need it. But I think right now, as an industry, fixed indexed annuities are going to show modest to slightly negative growth. And I'm perfectly fine with that. So we continue to encourage it, but we're not going to go out and go crazy with it because I think market conditions are just really tough right now. When you've got the 10-year running at 1.30%, I think it was when I last looked, I think that makes it tough. But we'll continue to encourage it because I think that trend will change. As respect to broker-dealer, the primary way for us to continue to drive that relationship is to encourage our insurance agents to get licensed. And so that remains a press for us. I think we've previously announced guidance that we want to do 1 in 5, we envision, of our agents. So the agent force continues to grow. We believe that, over time, roughly 1 in 5 should get a securities license. And today, we're at 1 in 7, give or take. So that's what we see as the opportunity, and that's how we're going about driving it.

That's helpful. A question for Paul. So you talked about the expenses, how 2019 saw some mandatory expenses that drove some of the increase, and you kind of alluded to, for 2020, you should see expenses to come down. Should we expect that in terms of the mandatory expenses to be largely flat year-over-year? And so by 2021, we could see even a further reduction on overall expenses in addition to your kind of planned expense saves?

Yes. So our expectation and built into our plan is that expenses will actually go down in 2020 versus 2019. I don't really want to go further than that in terms of guidance. I don't want to get into a reconciliation each quarter of specific dollars versus expectations. But you should expect broadly that expenses, particularly that primary bucket that's allocated to products but not individual product lines, will go down.

Ryan Krueger, KBW. On fee income, can you just give us any perspective on how you're thinking about the potential growth rate there going forward? You obviously had significant growth in the last year. But can you help us think about 2020?

Sure. So Ryan, again, I don't want to give specific guidance on that sort of component of earnings as well. I think sort of trending '19 into '20 would be appropriate adjusting for the change in assumptions that had the $6.5 million benefit in Q4 would be the best way to try to model it.

And then for Eric, the last couple of quarters, you've had the book yield go down about 5, 6 basis points. You gave us a 3- to 4 basis point example. I guess, is 3 to 4 basis points more your expectation going forward? Or is that just illustrative?

I think the 3 to 4 wasn't illustrative. But I wouldn't frame it, put it in the context of guidance or anything like that. But I think it's a model framework that you can use to develop your own view as to what to expect. I think I've pretty consistently over the last number of quarters, said given the puts and takes of the noise that flows through and all the adjustments that flow through the portfolio in a given quarter, anything from, I think I've said 3 to 8 kind of fits, is a model fit. And so all those the numbers we've just discussed would be within the realm, would square with my expectations. There are many, many other puts and takes than the ones we described here today.

Erik Bass with Autonomous. I wanted to follow-up on Humphrey's question on expenses. I think your guidance is to get to the $22 million of run rate savings by the end of this year. I guess, how quickly does that come in? And thinking about it, if you're getting to that run rate by the end of the year, it would imply kind of a lower level of expense in '21. And also, are there other expense savings that you kind of alluded to, but there may be other buckets in addition to kind of what you've outlined at this point that we should think about?

Sure. So Erik, I want -- I would encourage you to be careful about focusing too much on the examples that we've given. Because if you think about our direct expenses that are over $600 million, that $22 million is small. We focus on it because it's something specific that we've done that we can identify as an example of the kinds of things that we're doing to reduce expenses broadly. As I mentioned in my answer to Humphrey, I want to be very careful not to get into a scenario where we're reconciling to numbers every quarter. What I do want to leave you with is that we are being aggressive and thoughtful about how we manage expenses. Our intention and our expectation and it's built into our plan is that we reduce expenses. And that will support broadly earnings that are flat to slightly up in absolute dollars. And then through the benefit of share repurchase, achieve the 9% to 11% in our guidance.

Just one general comment to add to that. I think Paul touched on this. I just want to make sure I'm a little bit more forceful about it. We can't do anything about what's happening with interest rates and the impact on net investment income. But we also recognize that our shareholders expect us, fine, if you can't control certain things, you better go look at everything else that you can control. And that's the mind-set we're bringing. We have to do that every day, every quarter, we have to do that in this NII environment. We can't just let the NII control us. We have to do what we can do.

It's Alex Scott from Goldman Sachs. Question I had was just on the cash flow numbers that were mentioned. Should we consider the, I guess, what you've been doing in terms of capital behind new business, it was -- I just want to clarify, it was net of that? So that was what was it, $30 million to $50 million roughly? And then I also just wanted to inquire as to how much excess capital that assumes in terms of the drawdown? And I appreciate you don't want to give more detailed guidance, but just having that composition would help us better think about just ongoing regular way ex drawdown of excess capital, cash flow?

Yes. So maybe, Alex, just to clarify, when we talk about gross and net, what are we talking about? So for the last year or 2, we've referenced free cash flow before the capital needed to support growth. And that's basically assuming no growth. That when we say gross, that's really what we're referencing. What I think is more meaningful is what actually comes out of the Opcos up to the Holdco. And so the net free cash flow is what I'm referencing there. So the $290 million in 2019 was actually at Holdco and available for dividends, share repurchase and other growth opportunities. Well, largely M&A opportunities. The guidance that I've provided for 2020 is that same net concept at the Holdco, is between $330 million and $380 million. And my guess is you and others are doing the math and saying, well, wait a minute, if it was $330 million gross in '19 and you freed up $125 million, the $330 million to $380 million seems light. And it is based on that math. And obviously, the reason is we're hedging a little bit and retaining more in our model in the Opco for growth, potentially redeploy some capital in the asset portfolio, as Eric said, to the extent that -- the opportunities to take risk, and we feel we're being compensated, we could deploy some capital there.

Randy Binner, B. Riley, FBR. Just kind of a follow-up there. The sense -- is there a sensitivity in the lower RBC goal or guide to the move that we've had in interest rates? When you formulated this plan, the 10-year wasn't in the 1.30s. So is there kind of a quantum for every 25 basis points in the 10-year that we might want to not assume we get that cash out. Is there some kind of sensitivity you can provide on that?

Randy, thanks for the question. It's a good question. I'm just going to think out loud a little bit here. I don't -- I mean, the biggest exposure is to equities. And we actually saw that in the fourth quarter of '19. It had a big impact on our year-end -- sorry, fourth quarter of '18, had an impact on our year-end '18 RBC. Our exposure to equities is much lower today than it was back then in the wake of the up-in-quality trade that Eric talked about. So we don't have nearly the exposure there. In the fixed income portfolio, I mean, to the extent that the lower interest rates is causing stress for companies and the companies get downgraded, you could see an impact there, but I'm not sure, at least in the short term, there's a direct correlation. Eric, would you add anything?

Yes, Randy. As of today, I think we have an assumption for a certain amount of capital strain from assets that we build into the planning process every year. I don't see anything that threatens to turn that upside down at all right now. Paul mentioned, for example, that we have a substantially lower equity exposure or allocation than we did in 2018. It's about 1/10 of the size. So that's not going to do it. I think, while there have -- there's always going to be, if we have 900 CUSIPs, corporate CUSIPs, there's going to be a downgrade here or there. Some sparrows can fly into the window. I don't think any really big birds are going to fly into the window that are going to do any violence to the arithmetic Paul is talking about. So no. I think to the extent NII corrodes by a small amount in absolute dollar amount because of the interest rate environment, then, I guess, you might have that marginally small, less amount of stat income. Again, not going to do violence to the numbers that Paul is talking about.

Dan Bergman with Citi. Any help you can give us in terms of how to think about what level of interest rates is embedded in that 9% to 11% EPS growth guidance? If rates remain at current levels, is there some risk to hitting that range? Just any thoughts around how we should think about that would be helpful.

Sure. So in our 2020 plan, in the 2020 outlook that we shared, we presumed a new money rate of around 4%. With the 10-year sitting at 1.30% that might be a little bit of a stretch. But given that you get some of that back with spreads, we're probably not too far off. And my guess is we don't sit here at 1.30% forever. My, just my two cents is we end up settling out maybe at 1.50%. Again, with slightly wider spreads, we're probably not too far off 4% new money rates. We're also 2 months into the year. I think our new money rate for the fourth quarter will probably be at or around 4%. To try to dimensionalize it, if instead of for the full year, if instead of a 4% new money rate, it was 3.50%. That would reduce full year earnings by $6 million, $7 million pretax.

Ryan Krueger, I just -- separate follow-up. Maybe I'll get an answer from Karen, let you talk up there. On the indexed annuities, the 3 -- you have over a 3% spread, which is generally higher than I think most peers, at least that we see. I guess, are you still achieving similar spreads on new business at this point?

I mean, just like Gary said, we're always looking to see how we manage our spreads relative to the competitive environment. So we will get how we balance that out over the course of the year. But I would say, in general, yes, we're still looking to achieve those spreads. And I want to point out, too, that when you think about our spreads, you need to remember what I said before, which is, that's where the benefits expenses get paid out of and then that some portion of it gets dropped into our profit. But then generally, just like Gary said, we are looking to maintain our pricing discipline, even as we're tackling those interest rate headwinds.

Alex Scott from Goldman Sachs. I had one more follow-up. Just was interested, you mentioned the work you've been doing on LDTI. So just thought I'd ask if you have any updated comments you could provide on how you think that will change the way that we need to view your reporting. And I thought it was interesting that when you showed the asset liability durations that the assets actually had a little bit more duration than the liabilities. I think that probably is not the case for a lot of peers. So just thoughts around how that could affect what we kind of conclude as we get through all of that work and you guys disclose?

Sure. So maybe I'll start, Karen. And so on the LDTI, as I think I mentioned in my earlier comments, or maybe it was during the break, I don't recall. But in any event, we feel like we're ahead of the curve in terms of our planning and preparation for adoption in Q1 '22. We're not far enough along to have a view that we're prepared to share as to what the impact is upon adoption, and how to think differently about our financials on a quarterly basis post adoption. So that -- I think that's really all we can say on that topic unless you want more color around what we're doing to get prepared. On the ALM...

Yes, I think I would look to Eric to sort of talk about that, the fact that our assets are looking a little bit longer than our liabilities.

Yes. And that's worked well for us, as you can imagine, for a really long time, given the path of rates. It's also worked well for us because by the nature of having the captive distribution and the nature of the products we sell, they don't tend to be very interest-sensitive on the liability side. So the cash flows on the liability side are quite predictable. And that allows us to be thoughtful around duration matching. And we've been -- very consistently, we've applied pretty strenuous product level guidelines, where we don't allow ourselves very much tolerance at all for out -- being out of match with more interest-sensitive products in maybe some of the annuity type things, investment leverage type products. They're closely matched. You get into long-term care or a specified disease or universal life, some of the really longer tailed products. We tend to be a little more flexible. Because those cash flows happen a long time from now, and they're also, tend to be pretty consistent. So I think this is an area we thought pretty hard about. We have a very intentional process, and we're positioned very intentionally in that way.

Any other questions? Okay. We'd like to thank everyone very much for your attendance, both in person and online and look forward to speaking with you again soon.