NMI Holdings, Inc. (NMIH) Q4 2019 Earnings Report, Transcript and Summary

Ladies and gentlemen, thank you for standing by, and welcome to the NMI Holdings, Inc. Fourth Quarter 2019 Earnings Conference Call. [Operator Instructions]. I would now like to hand the conference over to your speaker today, John Swenson. Thank you, and please go ahead, sir.

Thank you. Good afternoon, and welcome to the 2019 fourth quarter conference call for National MI. I'm John Swenson, Vice President of Investor Relations and Treasury. Joining us on the call today are Brad Shuster, Executive Chairman; Claudia Merkle, CEO; Adam Pollitzer, our Chief Financial Officer; and Julie Norberg, our Controller. Financial results were released after the close of the market today. The press release may be accessed on NMI's website located at www.nationalmi.com under the Investors tab. During the course of this call, we may make comments about our expectations for the future. Actual results could differ materially from those contained in these forward-looking statements. Additional information about the factors that could cause actual results or trends to differ materially from those discussed on the call can be found on our website or through our regulatory filings with the SEC. If and to the extent the company makes forward-looking statements, we do not undertake any obligation to update those statements in the future in light of subsequent developments. Further, no one should rely on the fact that the guidance of such statements is current at any time other than the time of this call. Also note that on this call, we refer to certain non-GAAP measures. In today's press release and on our website, we've provided a reconciliation of these measures to the most comparable measures under GAAP. Now I'll turn the call over to Brad.

Thank you, John, and good afternoon, everyone. In the fourth quarter, National MI reported record results, capping a year of standout success. In 2019, we delivered exceptionally strong financial performance. Broad success in customer development and best-in-class growth in NIW and insurance-in-force. We continued to innovate in the reinsurance and capital markets and lead with discipline through Rate GPS in our individual risk underwriting approach. We delivered $182.4 million of full year adjusted net income in 2019, up 61% compared to 2018, and are exiting the year with a 23.3% fourth quarter adjusted return on equity. We activated 94 new customers in the year and generated over $45 billion of NIW, up 65% compared to 2018 We closed 2019 with $95 billion of high-quality and high-returning primary insurance-in-force. The investments we have made to develop and deploy a comprehensive credit risk management framework, spanning Rate GPS, individual risk underwriting and our innovative reinsurance program continue to drive value. In 2019, we delivered industry-leading credit performance with a full year loss ratio of 3.6%. Shifting to Washington matters. Last year, we welcomed new leadership at the FHFA, Fannie Mae and Freddie Mac. Our conversations with each and more broadly with other regulators and policymakers in Washington were constructive throughout the year. We are encouraged by the progress that was made in 2019 towards ensuring the long-term health of the mortgage market and expect these important efforts will continue in 2020. The mortgage insurance industry and National MI play a central role in the housing finance system, and we are pleased to see broad recognition in Washington of the value that MI provides to borrowers, lenders, taxpayers and others. Overall, I'm delighted with what we achieved last year, and I'm excited about our opportunity in 2020. With that, let me turn it over to Claudia.

Thanks, Brad. I'm pleased to report that in the fourth quarter, we once again achieved record performance, expanding our customer franchise, driving strong NIW volume and growth in our insurance-in-force and delivering record financial results, all while maintaining our disciplined approach to managing risk and return. GAAP net income for the quarter was $50.2 million or $0.71 per diluted share, and adjusted net income was $52.6 million or $0.75 per diluted share. GAAP return on equity was 22.3% for the quarter, and adjusted ROE was 23.3%. We generated record fourth quarter NIW of $11.9 billion, down seasonally from the third quarter of 2019, but up 72% compared to the fourth quarter of 2018. Primary insurance-in-force was $94.8 billion at quarter end, up 6% compared to the third quarter and 38% compared to the fourth quarter of 2018. In the fourth quarter, we activated 18 new lenders. For the full year 2019, we activated 94 lenders, including 6 from the top 200. We are now doing business with a broadly diverse group of nearly 1,100 high-quality originators. Our activation pipeline is healthy, and we expect to continue expanding our customer franchise in 2020. Equally as important, we are continuing to grow with our existing lenders. Leveraging our value proposition of certainty and service and our consultative approach to customer engagement to further strengthen our relationships. Rate GPS continues to be a standout success with our customers. More than 95% of our lenders are currently using the platform, and more than 90% of our fourth quarter NIW volume was delivered through the engine. The mortgage market is at an exciting point of change. Lenders are prioritizing technology to drive improvements in the consumer experience and streamline their business processes. As they do, their expectations for their mortgage insurance partners are evolving. They expect us to offer technology-enabled solutions to have embedded connectivity with third-party origination systems and point-of-sale platforms and to meet their drive for process efficiency with accelerated response times. In this context, Rate GPS and the broader technology lead that we enjoy provide a key advantage. We are meeting our customers' needs for speed and efficiency and expect to continue doing so as the mortgage landscape evolves. Our success with customers continues to drive strong growth in our insured portfolio. And our commitment to a broad-based risk management program spanning individual risk underwriting, Rate GPS and comprehensive reinsurance solutions continues to drive industry-leading loss performance. More broadly, private MI market conditions remain healthy, with low rates and the millennial demographic tailwind driving origination volume and consumer strength and rigorous underwriting standard driving favorable credit performance. Against this backdrop, we are delivering on our business plan and are excited about our record performance in the fourth quarter. We are focused on maintaining the right risk return balance and are confident in our ability to create significant shareholder value going forward. With that, let me turn it over to Adam.

Thank you, Claudia, and good afternoon, everyone. We had another strong quarter and achieved record results across a number of key financial metrics. We generated record fourth quarter NIW of $11.9 billion and continued the rapid growth of our high-quality insured portfolio. This drove record net premiums earned of $95.5 million, record adjusted net income of $52.6 million or $0.75 per diluted share and adjusted return on equity of 23.3%. Primary insurance-in-force was $94.8 billion at quarter end, up 6% from $89.7 billion at the end of the third quarter and up 38% compared to the fourth quarter of 2018. 12-month persistency in the primary portfolio was 77%, down from 82% in the third quarter. The current interest rate environment has helped drive incremental NIW into the MI market. However, it has also spurred increased turnover in the in-force portfolio due to refinancing activity. Assuming the current interest rate environment holds, we expect persistency will continue to trend down through the second quarter and then begin to rebound through the end of the year. Notwithstanding the increased turnover, we expect to continue delivering strong growth in our insurance-in-force as we go forward, given our view of future NIW potential. Total NIW was $11.9 billion, with monthly products contributing $11.1 billion or 93% of our total volume. Purchase originations represented 76% of our volume compared to 80% in the third quarter. Net premiums earned in the fourth quarter were $95.5 million, up 3% from the third quarter and 38% compared to the fourth quarter of 2018. We earned $8 million from the cancellation of single-premium policies compared to $7.4 million in the third quarter. Reported yield for the quarter was 41.4 basis points compared to 43.1 basis points in the third quarter. We expect net yield to trend down to between 35 to 37 basis points by the end of 2020, reflecting the strong growth of our NIW volume and focus on higher-quality risk cohorts, which carry lower premium rates, lower expected volatility and lower capital requirements along with the runoff of business that we originated and priced prior to the implementation of U.S. tax reform in 2018. Our yield guidance also reflects our expectations for reinsurance and ILN activity and the related impact on net premiums during the year. Overall, we continue to capture business at rates that are supportive of our strong mid-teens return objective. We continue to use Rate GPS to actively shape the credit mix of our new production and manage our concentration of business with layered risk characteristics. In the fourth quarter, our mix of greater than 45 DTI volume was 8.8%, and our concentration of 97 LTV and below 680 FICO volume were 5.5% and 2.1%, respectively, all well below the overall market. Investment income was $8 million, up from $7.9 million in the third quarter. Underwriting and operating expenses were $31.3 million compared to $32.3 million in the third quarter. In the fourth quarter, we have separately reported the third-party expenses incurred by our service subsidiary NMIS, which provides contract underwriting support on a fee-for-service basis to certain lenders. These service expenses relate to noninsurance activities and are generally offset in the same period by fees paid to NMIS, which we recognize as other revenue. Our service expenses and other revenue increased in 2019, along with the growth in our overall business and scaling of our customer relationships. We're focused on driving organizational efficiency at all times and actively work to leverage technology in a way that allows us to quickly respond to the needs of our customers and our rapidly growing business, while maintaining the smallest possible expense footprint. Our GAAP expense ratio was 32.8% compared to 35% in the third quarter and 42.4% in the fourth quarter of 2018. The calculation of GAAP expense ratio excludes NMIS revenue and service expenses as they relate to noninsurance activities. We had 1,448 notices of default in the primary portfolio at the end of the fourth quarter compared to 1,230 at the end of the third quarter. Claims expense was $4.3 million in the fourth quarter. Our fourth quarter loss ratio, defined as claims expense divided by net premiums earned was 4.5%. Credit remains strong, and our in-force portfolio continues to perform better than initially expected and priced. Overall, we expect to realize continued improvement in our combined ratio and underwriting margin expansion in 2020 as we further lever our fixed expense base and deliver strong credit performance in our insured portfolio. Interest expense in the quarter was $3 million, and we had a $2.6 million loss from the change in the fair value of our warrant liability. Moving to the bottom line. GAAP net income for the quarter was $50.2 million or $0.71 per diluted share. Adjusted net income was $52.6 million or $0.75 per diluted share compared to $49.9 million or $0.71 per diluted share in the third quarter and $32.1 million or $0.46 per diluted share in the fourth quarter of 2018. Year-on-year, we grew adjusted net income by more than 60%. We continue to organically grow our equity base and capital position at an accelerated pace. Shareholders' equity at the end of the fourth quarter was $930 million, equal to $13.61 per share, which compares to $873 million or $12.86 per share at the end of the third quarter and $701 million or $10.58 per share at the end of the fourth quarter of 2018. Year-over-year, we grew book value by more than 30%. Total cash and investments were $1.2 billion at quarter end, including $55 million of cash and investments at the holding company. In the fourth quarter, we received regulatory approval to allocate our holding company's stock-based compensation expense to our operating subsidiaries. Beginning in 2020, we will also have ordinary course dividend capacity available from our primary operating subsidiary for the first time. The ordinary course dividend capacity and allocation of stock-based compensation expense, meaningfully enhance the liquidity profile of the holding company and provide us with incremental financial flexibility going forward. At quarter end, total available assets under PMIERs grew to over $1 billion, which compares to risk-based required assets of $773 million. Excess available assets were $243 million at quarter end. In summary, we achieved record results in insurance-in-force, net premiums earned, expense ratio and adjusted net income and EPS. Looking forward, we believe that we are well positioned to continue delivering strong mid-teen returns that are significantly in excess of our cost of capital. We expect that the growing size and attractive credit profile of our insured portfolio, along with our broadly disciplined approach to risk management, expenses and capital optimization will continue to drive our performance. With that, I'll turn it over to Claudia for her closing remarks.

Thank you, Adam. Our record fourth quarter capped a year of sustained outperformance across our platform. For the year, we delivered industry-leading growth in NIW volume and insurance-in-force, best-in-class credit performance, significant expansion in our underwriting margin and return on equity and continued innovation in the reinsurance and capital markets. The broader economic environment remains strong, and private mortgage insurance market conditions are healthy. In this setting, we believe we are well positioned to continue to win with customers, drive growth in our high-quality insured portfolio, maintain the right risk return balance and deliver strong results for our shareholders. With that, I'll ask the operator to come back on, so we can take your questions.

[Operator Instructions]. And our first question comes from the line of Mark DeVries with Barclays.

It's still a little early but it looks like you may have lost a little bit of share in NIW in the quarter relative to peers. Anything to call out there on the quarter?

Yes. Mark, nothing specific to note. We continue to approach the business in the same way that's proven to be successful for us in the past. We feel really good about our results. We achieved $11.9 billion of NIW in Q4 and grew 72% year-over-year. So we're really proud of the team, and we're also excited about our opportunities in 2020. So nothing specific to note there.

Okay. And I'm sure you saw the recommendation from the White House today around the 10 bps increase in the G fee. Just curious on kind of what your expectations are around that, whether there'd be a recommendation for a similar increase in pricing at the FHA? And what the implications are longer-term for the competitiveness of GSE loans with private MI?

So Mark, this is Brad. So that's just in the first pass at the budget proposal and it's not the first time that, that kind of an increase has been proposed. 2 years ago, the very same increase was part of it and never was enacted. So we think it's very early days. And not likely to ultimately be enacted as proposed there. We think it's more likely that any kind of changes to the g-fees and LLPA framework will get considered and incorporated in an overall GSE reform package, which gets amount of conservatorship.

And our next question comes from the line of Phil Stefano with Deutsche Bank.

I was hoping to talk about the adverse development in the quarter. It's something to feel it's out of trend with peers, and I think even there was a comment in the prepared remarks that the business continues to perform better than expected. So maybe, you can comment on what drove the adverse development? And any thoughts you could provide to help us get comfortable that this is something that we should be worried about persisting.

Yes. So I think it's a bit mechanical. So it's important to note, we don't have a legacy set of loan exposures, like many of our peers do. And so that means we don't have a population of borrowers that will likely to serially default cure and then default again. Ours is a much cleaner profile. And it is a resoundingly high-quality insured portfolio, the outcomes that we're achieving now in terms of default rate experience, claims experience, loss ratios are meaningfully better than the assumptions we embedded when we priced the business. But we will naturally have some amount of loss, that's why we're in business. It's to absorb these types of events. And as a general rule, we should see an amount of strengthening in our prior period reserves because we only book reserves once a borrower has defaulted. We're not allowed and don't book a loan loss provision for anticipated defaults like lenders do. So we book a reserve once we have a reported delinquency and then we increase the reserve as the delinquency ages. For us, alone, that's 180 days, past due will carry a higher reserve charge than a loan at 60 days past due. And so in any given period, we will see a progression of some portion of our default population to older age status. Now sometimes, that will be -- and there will be an increase in the reserve, that's a result of that. There will be periods where that's offset fully by cure activity on other previously reported defaults or modest changes in our underlying frequency and severity assumptions. And in other periods, that's just not going to be the case. That's what came through in Q4, is -- there was a significant amount of cure activity. We generally see cure activity seasonally dips in the fourth quarter as well. But it's just that natural dynamic of the aging of our portfolio and the aging of the defaults that we have that are previously reported. Without that offsetting, population of serial redefaulters cures and then defaulting again.

Is the seasonality of Q4 would -- I guess, I noticed the last 2 quarters that there was adverse development was fourth quarter '15 and '16, I believe. Is the seasonality help explain? Or is there something different in the reserving process in Q4?

No. There's nothing different in the reserving process in Q4. If you rewind the tape and you look at '18 and '17, those numbers are going to be a bit distorted by the hurricane activity that was coming through. But there is nothing unique about how we approach reserve setting in the fourth quarter of '19 or differences in the seasonal dynamic that we saw.

Okay. And maybe, switching gears, and maybe this is a thinly veiled NIW question. I don't -- but has there been a change in the conversations as you do more business with current customers, sign up new customers or was there some kind of change in the risk-based pricing algorithm that may have translated to an NIW differently than we were expecting?

No, Phil. There hasn't been a change. We always look at our risk-adjusted return when we're pricing. I think as far as just the momentum and the changes, it's -- the drivers are really good here. And we're doing the same things we've done in the past. There's nothing that really is a different approach. We're still looking at signing on new customers, building wallet share. The drivers for all of this in 2020 are good for us. We've got a really strong response from a customer level. We've got a focused strategy with Rate GPS, a compelling value prop. And the backdrop is healthy of MI market that remains really healthy with low mortgage rates and combined with low unemployment. So nothing we're doing differently. We continue to do with what we've done in the past, which makes us very successful.

And our next question comes from the line of Rich Shane with JPMorgan.

When we look at the development of the balance sheet, no matter how you want to look at it. Let's focus on risk in-force and we look at the formation of capital. 2020 is likely to be an inflection point where the formation of capital is growing at a faster rate than the use of capital. Are we approaching the point where you guys are considering returning capital? And how would you consider that?

Yes. Rick, it's a great question. We're not approaching that point. I think we've got a significant amount of capital runway and certainly have increased flexibility at the holding company, given our emerging opco dividend capacity in this SBC allocation, the stock-based comp allocation. But as we think about it, our customer franchise and NIW are expanding every day. And our in-force portfolio is growing at a rapid pace. So our capital needs are also growing as we roll forward. Altogether, we think, one, that we will need the capital that we're not at a point of being capital self-sufficient and that won't emerge in 2020; and 2, we think when we're delivering consistent plus 20% returns on equity that it's a valuable time to be continuing to invest in the business. And so we don't have plans today to initiate a common dividend or repurchase program. Certainly, over the long term, we need to think about the best ways to either deploy capital, invest in the business or distribute it. That's not going to be a 2020 focus item for us.

And our next question comes from the line of Bose George with KBW.

Actually, I wanted to ask about the guidance you gave on the premiums of 35 to 37 basis points. Does that assume that the contribution from cancellations declines in the back half of the year as refi activity slows?

Yes, both. It does. So it assumes a few things, and I think it's worthwhile to go through. We do assume that we're going to have some pressure on what we call our core yield, given the combination of the significant NIW that we're generating today and its quality, right, with quality, lower risk, lower volatility, lower capital that brings a lower premium rate. And also the runoff of some of those earlier vintages that had very high premium rates because they were priced prior to the effectiveness of U.S. tax reform. So there's a bit of it that relates to our core yield. We do have an expectation for certain activity in both the ILN and the reinsurance market in 2020, and that's included. And then we do, also, anticipate that we'll see a reduction in the contribution from our cancellation earnings. In terms of that dynamic, look, we have a growing pool of insurance-in-force. Even if we had a static dollar contribution from cancellation earnings, that would represent a smaller percentage yield component of the insured portfolio. And if you look at how we're entering 2020, we exited 2019 with $137 million unearned premium reserve balance, whereas we ended 2018 -- started 2019 with $159 million unearned premium reserve balance. That's the balance, right? It's the acceleration of those deferred premiums that come through upon cancellation of a single premium policy. So we expect a lower contribution in terms of the yield component from cancellations. We also actually are anticipating that we'll see a smaller dollar amount of cancellation earnings coming through in 2020.

Okay. Great. That's helpful. And then, just based on current pricing, I mean, is it -- can you tell in 2021, do you see further pressure on the premiums? Or is it just too early to tell on that?

Yes. It's a great question. I'd say, we generally don't provide longer-term guidance. And ultimately, our long-term yield is going to depend on a whole bunch of things, right? Our NIW volume, the risk profile of our new production and so by extension, the rate on our new business and then the persistency of the in-force portfolio, and also, what we choose to do in the reinsurance market. And so it's really just too early for us to give you a steer on where yield is likely to head in 2021 or beyond.

Okay. Fair enough. And actually, just one on the tax rate. What drove the slightly lower tax rate? And then can you just remind us the sort of normalized tax rate next year?

Yes. It's a good question. So starting point for us, we've got a 21% statutory rate. And when we establish our provision and also when we provide guidance, we sort of start with that 21% rate. We scale it down for source of the revenue that are excludable like investment income on muni portfolio and we scale it up for certain expenses that aren't deductible. And so on balance, that will have us at around 22% to 23% estimated effective rate. Our actual tax rate in any given quarter, though, reflects the impact of certain, what we call, discrete items that occur in a particular period. The 2 most notable items that we've highlighted in the past, are fluctuations in the fair value of our warrant liability and equity awards, the vesting of RSUs and exercise of options. We had some equity activity come through as a discrete item in Q4. And so we saw an amount of options being exercised, and that had a positive impact on our effective tax rate. You may recall that, that's typically what we see come through in the first quarter of each year. And so as we look into 2020, we would expect to see a full year effective tax rate of around 22% but we expect our Q1 rate to be a bit lower than that at around 20%. And we expect our Q2, 3 and 4 tax rates to be modestly above our full year guidance of 22%.

And our next question comes from the line of Douglas Harter with Crédit Suisse.

Can you talk a little bit more guidance around expenses? How to think about the overall dollars of expense growth as the insurance-in-force is still growing at a healthy clip?

Yes. Doug, it's a great question. Let me offer you a bit of a steer for 2020. I'd say, overall, right, it's incumbent upon us, and we always look to balance investments in things like technology, staff additions, other items where we think we need to grow. Balance those against really the need for us to maintain expense discipline. And we think today that we are striking that right balance. We're fully supporting our business needs, investing and supporting the growth that we're seeing, while at the same time, levering into our operating expense base at an accelerating pace. And so we see that coming through, right, and the improvements in our expense ratio. In 2020, we expect that dynamic to continue. We expect to continue levering into our fixed expense base, and driving expense ratio improvement. I'd say, overall, we anticipate that adjusted operating expenses, and I'm using that term adjusted operating expenses pointedly, will grow by 10% or less compared to 2019. In terms of how that plays out sequentially through the year, quarter-to-quarter, we typically see modestly higher expenses in Q1 than we do through the remainder of the year because we reset certain employee benefit items like our 401(k) match and payroll taxes. And so that tends to drive slightly higher expenses in Q1 than we then see developing through the remainder of the year.

And our next question comes from the line of Mark Hughes with SunTrust.

On the persistency, how much further down, do you think it will go? And what is the -- what drives that? How -- what are the timing impacts, when you see a lot of the refi activity, a lot of turnover in the market?

Yes. So I'll answer both of those in turn, Mark. So in terms of how low it will go, we're not going to be providing guidance, specific guidance. We do expect that our 12-month persistency will dip below the 77% that we reported in Q4 but then rebound in the second half of the year. If we tied it with that, we do certainly expect to continue delivering industry-leading growth in our insurance-in-force, given our expectations for future NIW potential. In terms of the dynamic that drives this, it's really related to the persistency calculation itself. So our reported 12-month persistency, looks at the total primary, if, that was on our books 12 months ago. So in this instance, at December 31, 2018, and look at how much of that insurance-in-force remains in the portfolio 12 months later. In this instance, at December 31, 2019. And so the set we reported 77% persistency. That persistency ratio or percentage doesn't consider any of the business that we wrote in 2019 since that business wasn't in the portfolio 12 months ago. And so our persistency is heavily influenced by a few things, right? It's heavily influenced by the profile of the business that we wrote, right? What's the underlying note rate but also when we wrote it. And so what we expect to see in 2020 is that as more and more of the high-quality, but we expect to be highly persistent business that we wrote in 2019, comes into that calculation, right? So as of 3/31/20, our reported 12-month persistency will now include the first quarter 2019 production. Roll that forward to the second quarter, will begin to include the second quarter '19 production. And if you look at how note rates have developed through 2019, we expect the business that we originated, really beginning in the second and then on in the third and fourth quarters, because of the underlying node rates to be incredibly persistent. As that starts to come into the calculation, we'll expect to see our persistency ratio rise.

And our next question comes from the line of Jack Micenko with SIG.

Adam, I'm curious to understand how you're thinking about the sensitivities on the premium yield on a 37, 38 versus like a 34, 35. And I know you got it to 35 to 37. But is that just ILN timing? Or are you thinking that competition, maybe, gets a bit more pronounced in that total range you've given yourself?

No. As a general matter, we see the broad rate environment is being generally stable today, right? There's a few small pockets where we've seen competition in the same ways that we've seen for some time now. But we are still pricing all of our business to achieve the same risk-adjusted return of 15-plus percent risk-adjusted return. Importantly for us, on an unlevered PMIERs basis. No, we don't consider the source or cost of the capital that supports that. We really are pricing as, though, we're fully loading with equity. And we don't see that changing because we're not observing anything in the market today that causes us to think that dynamic is going to shift. In terms of the outcomes that we'll see during the year but there are always going to be uncertainties, right? Where does persistency trend exactly? What happens to the in-force portfolio? What happens really by extension with the broader interest rate environment? Where does cancellation contribution come through? And then, also, what our risk appetite is, right? Our risk appetite and by extension, our rate is not going to be static. And so all of those things are the items that we expect will drive different outcomes in the reported yield. And then as you noted, maybe, exact timing and also the terms that we achieve on the reinsurance transactions that we expect to pursue this year. All of those will come through. It doesn't contemplate changes in the broader pricing environment because we're just not seeing that today.

All right. And then the 2018 loss incurred vintage to date numbers are looking like they're trending higher than prior vintages for you and also Radian reported a similar trend. I'm curious why you think that is? That the '18's are hitting higher loss incurred sooner than prior vintages?

Yes. Look, I would say, broadly speaking, these are low -- small numbers, right? And so we think the 2018 vintage is going to perform in an exceptionally strong way for us. When we model out lifetime losses, we're still anticipating meaningfully better credit performance than what our models and price assumptions were at the time we originated that business. Also, for us, we've got that book completely covered by reinsurance, both quota share and ILN cover. So we're really comfortable with what that means as we progress forward. In terms of the specifics, I think, there's 2 items that play. One is, if you look at the underlying profile of that business, that's when it was really beginning in late '17, but in a more pronounced way in early '18 that we saw a proliferation of some higher risk business coming through the market, right? The increase in the concentration of high DTI borrowers coming through with some suboptimal layer risk characteristics. That was the genesis of Rate GPS. And we rolled out Rate GPS on June 4, 2018. We don't actually start getting NIW through the engine. We get apps, we get commitments but we don't get NIW until a few months later. So a large chunk of our 2018 production was originated on a rate card basis, and we recognized that the rate card wasn't the perfect tool at that time to address the risks that were in that market. I think you are seeing a validation of our decision to introduce Rate GPS in how that book is performing relative to some others. And also, we saw, when we had a rise in interest rates as we went through the back end of '18, we had a plateau and a little bit of a dip in the pace of home price appreciation. And so -- now that has reaccelerated. So we'll expect things to broadly speak and get better. So when you have a temporary dip in HPA, particularly in the first year or two after you've originated a book of business, you do expect that to come through into performance, certainly relative to the HPA tailwinds that the 2016 and '17 vintages had in their early years.

And our next question comes from the line of Chris Gamaitoni with the Compass Point.

Most have been answered. Adam, what did you mean by adjusted operating expenses, when you kind of gave your outlook of 10% or less?

Sure. So I mean, two things. One, I want to make sure that you're hearing how we are reporting our operating expenses beginning in the fourth quarter, which excludes the third-party service expenses from NMIS and then also, you'll recall, when we report our earnings, we -- one of the items that we adjust for that we disclosed is transaction-related expenses. And so in 2019, we had about $2.4 million of transaction-related expenses. So if you look at our total GAAP expense profile and your models may not be tuned yet to exclude service expenses. So if you just look at the total GAAP expenses, back out the service expenses, back out the transaction costs, that's the point of reference that I'm talking about growing 10% or less. We will then separately expect to have transaction-related expenses for the ILN activity that we may pursue in 2020. And we will also separately have NMIS third-party service expenses, but that's scaling, I want to make sure you're looking at the right starting point.

Okay. Sure. And does that adjusted number include the impact of the ceding commission?

It does.

Okay. And then I realize the capital consumptive nature of the business, still. I was just wondering if you had any color on what the type of dividend capacity is now just so we think about flexibility potential in the future?

Yes. So I think about it in concert with our stock-based comp allocation. The reason that I highlight that, we have a really, really robust and rich expense sharing arrangement between the holding company and the operating company expenses that the holding company incurs for the benefit of the operating company. We've got agreements in place that allow the allocation. Stock-based compensation, though, is a noncash expense. And so the ability to allocate a noncash expense from the holdco to the opco and have that settled with a cash payment up from the opco, increases the liquidity flow, sort of, the free cash -- available free cash flow, I'll call it, from the operating company. The opco dividend capacity and the stock-based comp allocation, we'd expect in 2020 to yield about a $25 million potential free cash flow, whether we will extract the component of that, that relates to the stock-based compensation that's an expense item that we will, along with all of the expense items that we've been allocating, we'll do. The operating company dividend is about, in total, between NMIC and Re One will be a little bit over $16 million. That's not something that we will necessarily extract but it's available to the extent that we wanted to.

And just -- whether -- now that's stock-based comp is in cash but kind of, what are the cash expenses out of the holdco, rough numbers, right?

You mean, cash flow out, Chris or...

Any cash flow expenses that have to be paid out of the parent, the holding company?

Yes. That's essentially neutral now, Chris.

And our next question comes from the line of Randy Binner with B. Riley FBR.

I had a couple. Just that service fee expense item that was excluded, is that an item that kind of grows with the size of the company? Or is that pretty static at around $1 million a quarter? Is that the right figure?

Yes. Yes, so it will grow. It's not necessarily just with the size of our company but it's going to grow, depending on which customers we engage with. Certain lenders utilize contract underwriting services more than others. And also, it depends on the refi environment, it tends to be higher in refi heavy environment, that's when lenders need contract underwriting support. From forecasting standpoint and what it looks like going forward, I would guide you, I don't want to say it doesn't matter, It's a business that we invest in that we offer for our customers that drives value for us. But overall, we expect that any growth in NMIS service expenses will be paired and fully offset by growth in our other revenue contribution because we're getting fees. We're getting reimbursed for the incurrence of those expenses.

Got it. And then just on reinsurance, the use of both quota share and then the ILN forms of reinsurance. It is impacting the printing yield, a little bit but it's also incredibly efficient source of capital support. So how much of the business do you see in the next 1, 3, 5 years being kind of syndicated out to the reinsurance market?

Yes. Look, I see a significant majority of the business being syndicated out to the reinsurance markets broadly over the next several years. It's important for us. We talk about as a strategy matter, having a comprehensive risk management program and a core pillar of that risk management program is the use of reinsurance structures. There's obviously a lot more that we do on the front end with individual underwriting and Rate GPS. But it's not just as a funding matter but also as a risk matter, we see great value. But reinsurance for us, in all of its forms, gives us -- we think, it's better than debt-like cost of funding, right, sub-5%, in many instances, far below 5%. With equity-like loss absorption characteristics. And those loss absorption characteristics, in particular, are expensed as we get out into sort of theoretical tail events. And so we will want to have reinsurance structures in place. Now the exact contours of our reinsurance program may change over time. We may not always have a quota share program. We may want to evaluate a traditional excess of loss program and to the extent that we could do something that was both additive from a cost of capital standpoint and actually enhanced our risk protection, we'll look to do those things. And so it's not just going to be a cut and paste uniform impact on our yield sort of uniform cost that gets scaled up for the growth in our book. We are going to be nuanced in how we utilize those markets. But I do see them as being central to what we do going forward.

Well. Yes, I appreciate that. And I agree, it's a good use of capital and could argue for a higher multiple in the stocks. Just kind of how are those negotiations and ongoing conversations going? Is that -- is the reinsurance market still broadly available from a capacity perspective?

Yes, it is. We're working on something now in the traditional market. We're not at a point where we're going to be disclosing anything but we're having some good conversations. We'll see where we land. And then more broadly, as we look out at the ILN market, in particular, we would generally observe that the ILN market has offered really favorable terms to issuers with objectively worse quality of reference pools so far this year. And so we're optimistic that we'll be able to, once again, positively differentiate when we come to that market, when we're ready to transact in 2020.

And our last question comes from the line of Geoffrey Dunn with the Dowling & Partners.

My questions have been asked and answered.

And that does conclude today's question-and-answer session. I would now like to turn the call back to management for closing remarks.

Thank you for joining us on the call today. We'll be at the Crédit Suisse Financial Services Forum on February 27 and the RBC Global Financial Institution's conference on March 10. We look forward to seeing you then.

Ladies and gentlemen, this concludes today's conference call. Thank you for participating. You may now disconnect.