Radian Group Inc (RDN) 2020 Q4 法說會逐字稿

Welcome to the Radian Fourth Quarter 2020 Earnings Call. My name is Annie, and I'll be your operator for today's call. (Operator Instructions) Please note that this conference is being recorded.

I will now turn the call over to John Damian, Head of Investor Relations and Corporate Development. Mr. Damian, you may begin.

Thank you, and welcome to Radian's Fourth Quarter and Year-End 2020 Conference Call. Our press release, which contains Radian's financial results for the quarter and the year was issued yesterday evening and is posted to the Investors section of our website at www.radian.com. This press release includes certain non-GAAP measures, which will be discussed during today's call, including adjusted pretax operating income, adjusted diluted net operating income per share, adjusted net operating return on equity and real estate adjusted EBITDA. A complete description of these measures and the reconciliation to GAAP may be found in press release exhibits F&G and on the Investors section of our website. In addition, our related non-GAAP measure, real estate adjusted EBITDA margin is calculated by dividing real estate adjusted EBITDA by GAAP total revenue for the real estate segment.

This morning, you will hear from Rick Thornberry, Radian's Chief Executive Officer; and Frank Hall, Chief Financial Officer. Also on hand for the Q&A portion of the call is Derek Brummer, President of Radian Mortgage. Due to the current environment, all of our speakers are remote. I would ask that you please excuse any sound quality or technical issues that may arise during the call.

Before we begin, I would like to remind you that comments made during this call will include forward-looking statements. These statements are based on current expectations, estimates, projections and assumptions that are subject to risks and uncertainties, which may cause actual results to differ materially. For a discussion of these risks, please review the cautionary statements regarding forward-looking statements included in our earnings release and the risk factors included in our 2019 Form 10-K as updated in our quarterly report on Form 10-Q for the third quarter of 2020 and subsequent reports filed with the SEC. These are also available on our website.

Now I would like to turn the call over to Rick.

Thank you, John, and good morning. Thank you all for joining us today and for your interest in Radian. As we look back at last year, it is safe to say that 2020 did not play out as we had originally planned. In March, the COVID-19 pandemic triggered a global health and economic crisis and caused an abrupt shift in our day-to-day focus. Combining a pandemic was heightened social unrest and a divisive political environment, I think it's fair to say that 2020 was a perfect storm. Despite the challenges of 2020, including shifting to a largely virtual work environment, we were able to write record-breaking levels of new mortgage insurance business and grow revenues in our real estate segment. And while our quarterly and full year results were impacted by the environment, I'm proud to say that our business model, weathered the storm as designed, demonstrating that through the cycle resiliency we and the mortgage industry have been building since the last financial crisis. I'm pleased with our ability to operate well with strong momentum throughout a challenging year. Although we continue to navigate the pandemic economic environment, as we enter 2021, we are encouraged by the signs of recovery and improvement in the overall economy, the continued strength and momentum in the housing market, and the positive default trends in our portfolio. I'd like to recognize our team across Radian and thank our customers, investors, business partners and Board for their commitment and support in helping us deliver solid results during this unprecedented time.

Frank will discuss the details of our financial position shortly, but let me first share a few highlights for 2020. We wrote nearly $30 billion of NIW in the fourth quarter, which is a 49% increase over the fourth quarter of 2019. This contributed to our record-breaking volume of new flow business written in 2020 of $105 billion, which represented a 47% increase year-over-year and marked our fifth consecutive year of record annual volume. We grew our primary insurance in force to $246 million, while the high-volume of refinances during year resulted in lower persistency. It's important to note that our high-quality insurance portfolio grew approximately 2% year-over-year, and our monthly premium insurance in force grew 11% year-over-year. Our mortgage insurance portfolio, which is one of the largest in our industry, is the primary driver of future earnings for our company. The economic value and the projected future earnings of this portfolio include the addition of the high-quality 2020 vintage, which represented more than 40% of our insurance in force as of December 31, 2020. We grew our book value per share by more than 11%. We achieved this growth even after accounting for nearly $100 million of dividends that we returned to stockholders in 2020 through the significant increase to our quarterly cash dividend in the first quarter.

In our mortgage segment, we continued our focus on meeting the needs of our customers and maximizing the economic value and future earnings of our mortgage insurance portfolio. Given the flexibility of our Radar Rates pricing model, we were able to respond to the rapidly changing economic environment in 2020. Our strong customer relationships and excellent service delivery, combined with market intelligence and value-based pricing decisions, helped our team effectively navigate the competitive environment to originate a high-quality book of business with significant economic value. Written at historically low interest rates, which should benefit persistency, we expect the 2020 vintage to produce attractive returns and contribute significant earnings in future periods.

For our real estate segment, despite a challenging pandemic environment, total revenues for the full year were $102 million, a 14% increase compared to 2019. While certain of our businesses experienced slowdowns, starting 2020 as a result of the pandemic, we saw strong growth in our title business, including a 238% year-over-year increase in closed title insurance orders with a strong sales pipeline of large customers going into 2021. The operating loss in the real estate segment for the fourth quarter and the year was largely a result of slowdowns in our valuation and REO businesses resulting from the foreclosure in a moratoria, staffing up to support our growing title business and our continued strategic investment in data analytics and technology across our digital valuation and real estate business platforms. We remain focused on positioning our real estate businesses for future growth by driving competitively differentiated data-driven digital products and services to our customers, lenders, realtors, consumers and mortgage investors. Although these real estate businesses are in the early stages of their development and maturity, we are very excited about the future and remain confident that we are well positioned with the customer relationships and an experienced team to drive increased stockholder value going forward. We will share more details about these businesses as we progress, including during our next Investor Day, which we plan to announce for later this year.

In 2020, we took steps to fortify our capital position to increase our strategic financial flexibility. During the second quarter, we strengthened our available liquidity by extending the term of our existing credit facility to January of 2022 and issuing $525 million of senior notes due 2025. We continue to execute our aggregate, manage and distribute business model focused on lowering the risk profile and through the volatility of the business. In 2020, we entered into a single premium quota share reinsurance program, which covers the 2020 and 2021 vintages of our single premium production and executed 2 mortgage insurance notes transaction for a total of $878 million.

At December 31, Radian Group maintained a strong capital position with $1.4 billion of total holding company liquidity and Radian Guaranty's PMIERs excess available assets grew 38% to more than $1.3 billion during the fourth quarter of 2020. I am pleased with our ability to strengthen our capital position and increase our financial flexibility, a true testament of our through-the-cycle business model. We believe the impact of the pandemic has been on earnings, not a capital event for the industry, and that Radian is well positioned to weather the remaining uncertainty ahead with our strong PMIERs capital and holding company available liquidity.

Moving now to the broader mortgage and real estate market. We continue to see this market perform extremely well through the pandemic environment, with strong purchase volume growth and significant home price appreciation. As we have all redefined the significance of home in so many aspects of our lives during the pandemic. The housing market flourished in 2020, driven by low interest rates and strong demand. Positive momentum continued in the fourth quarter with December existing home sales increasing 1% from the prior month and 23% from December 2019. And based on the latest data from our own Radian home price index over the last 3 months of 2020, strong housing demand and relatively limited supply in the market led to an annualized 9% increase in home prices across the country. We expect the rate of home price appreciation to moderate this year and believe the combination of continued strong supply and demand dynamics, low interest rates and income growth are well aligned for a healthy and sustainable housing market. Looking ahead, we expect the vaccine rollout and government support to sustain continued improvements in the economy and U.S. housing market and anticipate continued growth in home purchase activity and gradual reductions in refinances.

For 2021, recent market projections estimate total mortgage originations of approximately $3 trillion. While the overall origination market is expected to be smaller in 2021 as compared to last year and expectations for refinance volume vary, there is consensus around a growing purchase market in 2021, which is positive for our industry, given the higher likelihood of purchase loans to utilize private mortgage insurance as compared to a refinance loan. Based on these projections, the private mortgage insurance market is expected to be approximately $450 billion to $500 billion. It is also important to note that as refinance volume declines, we benefit from increased persistency in our portfolio, further supporting insurance support growth.

Turning to the regulatory legislative landscape, the Biden demonstration has laid out an ambitious plan for housing, focused on increasing access and affordability for low and moderate income borrowers, increasing the inventory of affordable homes and ensuring an equitable housing finance system. We support the administration's focus in these areas, which are aligned with our corporate mission of ensuring a pathway to responsible, sustainable homeownership. With the change in administration, there's been increased discussion of a potential FHA premium reduction. Based on our recent reports about a possible 25 basis point decrease, we would anticipate a small impact to our industry and to our business volume. The government continues to provide broad support to homeowners impacted by the pandemic. The recent announcement by the FHFA of an extension of GSE mortgage forbearance from 12 to 15 months should help to further assist borrowers through the crisis. In the unprecedented level of government stimulus, including the new relief package that is expected to pass Congress shortly, should help accelerate the economic recovery already underway. We are very proud of the role that private mortgage insurance industry has served throughout this crisis, aligning our policies and procedures with the GSEs to support struggling homeowners, while at the same time, continuing to serve our critical role of taking on first loss mortgage credit risk in one of the largest housing markets in history.

Finally, as we've said before, our industry is the only committed source of permanent private capital for U.S. mortgage credit risk and has continued to consistently underwrite and support mortgage credit risk through the market cycles. Our overall performance during the pandemic, which has been a period of extreme economic stress as well as our ability to continue to effectively manage and distribute risk through the cycle to further strengthen our company has proven the resilience of our business model and capital structure.

Now I would like to turn the call over to Frank for details of our financial position.

Thank you, Rick, and good morning, everyone. To recap our financial results issued last evening, we reported GAAP net income of $148 million or $0.76 per diluted share for the fourth quarter of 2020 as compared to net income of $0.70 per diluted share in the third quarter of 2020 and net income of $0.79 per diluted share in the fourth quarter of 2019. Adjusted diluted net operating income was $0.69 per share in the fourth quarter of 2020 as compared to adjusted diluted net operating income per share of $0.59 in the third quarter of 2020 and adjusted diluted net operating income per share of $0.86 in the fourth quarter of 2019.

I'll now turn to the key drivers of our revenue. As Rick mentioned earlier, our new insurance written was $29.8 billion during the quarter compared to $33.3 billion in the third quarter of 2020 and $20 billion in the fourth quarter of 2019. New insurance written for refinances was 35% of total new insurance written for the fourth quarter of 2020 compared to 30% in the third quarter of 2020 and 33% for the same quarter in the prior year. Direct monthly and other recurring premium policies were 91% of our new insurance written this quarter, a slight increase from 90% for the third quarter of 2020 and 82% for the fourth quarter a year ago, which also means that single premium policies were down significantly to only 9% of our quarterly new business. In total, borrower paid policies were 99% of our new business for the fourth quarter.

Primary insurance in force increased to $246.1 billion at the end of the quarter as compared to $245.5 billion in the third quarter of 2020, with year-over-year insurance in force growth of approximately 2%. As a reminder, last quarter, in addition to elevated policy cancellations due to the current low interest rate environment, we also experienced additional single premium policy cancellations during both the third and fourth quarters as part of our ongoing servicer monitoring and reconciliation process. These additional cancellations represented approximately $1.8 billion of insurance in force in the fourth quarter of 2020 and $2.9 billion in the third quarter for a total of $4.7 billion. Absent these cancellations, insurance in force growth would have been approximately 4% year-over-year, and it is important to note that monthly premium insurance in force, which was not materially affected by these cancellations, has grown 11% year-over-year. And the overall mix of our in-force portfolio has shifted from 28% singles a year ago to 22% as of year-end 2020.

Our 12-month persistency rate of 61.2% decreased from 65.6% in the prior quarter and 78.2% in the fourth quarter of 2019. Our quarterly annualized persistency rate was 60.4% this quarter, a slight increase from 60% in the third quarter of 2020, and a decrease from 75% in the fourth quarter of 2019. The year-over-year decline in quarterly annualized persistency is primarily driven by the continued high level of refinance activity. Given the current mortgage rate environment, it is expected that near-term persistency will remain below our expected long-term trends.

Moving now to our earned premiums. Net premiums earned were $302.1 million in the fourth quarter of 2020 compared to $286.5 million in the third quarter of 2020 and $301.5 million in the fourth quarter of 2019. The increase of 5% on a linked-quarter basis is primarily driven by $11.3 million related to changes in present value estimates for initial premiums on monthly mortgage insurance policies that are deferred but not collected until cancellation. And the impact of a line item reclassification related to our title insurance business recorded in the fourth quarter to adjust earlier periods in 2020, which increased net premiums earned and decreased services revenue by $7.8 million, further details of which are presented on exhibit E.

Slide 11 shows the mortgage insurance premium yield trend over the past 5 quarters, excluding the impact on both the previously noted $11.3 million adjustment in the fourth quarter as well as a similar $17.4 million adjustment in the fourth quarter of 2019. Our direct in-force premium yield, as noted on Slide 11, was 42.8 basis points this quarter compared to 43.2 basis points last quarter and 47.1 basis points in the fourth quarter of 2019. As noted in previous quarters, we expect our in-force portfolio yield to continue to decline due to the difference in credit mix and associated premium rates of today's NIW relative to prior vintages. Recent trends of lower persistency and higher levels of new insurance written have also contributed to a faster rate of change in the yield of our mortgage insurance portfolio. The timing and magnitude of future portfolio yield changes will continue to depend on several factors, including the volume, mix and pricing of new business relative to volume and mix of cancellations and prepayments in our portfolio. Our level of premium yields driven by single premium cancellations was 8.7 basis points compared to 10.7 basis points in the third quarter of 2020 and 4.4 basis points of yield in the same quarter a year ago. On a linked-quarter basis, the decline in cancellations associated with our ongoing servicer reconciliation process is the primary driver of the decrease. Approximately 1.5 basis points of the gross yield related to cancellations was due to the previously mentioned servicer reconciliation activity that occurred in the quarter.

With regard to pricing on new business, we remain focused on maximizing economic value and generating attractive risk-adjusted returns, which we target at between 13% to 17%. These targeted returns do not include the impact of insurance-linked notes transactions but do incorporate the impact of our single premium quota share reinsurance program, which is a forward commitment by our panel of reinsurers and is in place at the time of loan origination. Real estate segment revenues were $23.6 million for the fourth quarter of 2020, representing a 21% decrease compared to $29.8 million for the third quarter of 2020 and a 7% increase compared to $22 million from the fourth quarter of 2019. Our reported real estate adjusted EBITDA for the fourth quarter of 2020 was a loss of $7 million. The decrease in real estate adjusted EBITDA in the fourth quarter and full year 2020 compared to the fourth quarter of 2020 and full year 2019 was primarily related to the negative impact of the COVID-19 pandemic on the operating environment for certain of our businesses and our continued strategic investment in growing our title and digital valuation in real estate businesses.

Due to certain changes that we made in the composition of our reportable segments in the fourth quarter of 2020, our results for both the real estate segments and the all other category have been restated for all prior periods to reflect these changes. See Exhibit E for more details on these reclassifications. And finally, our investment income this quarter of $38 million was up 5% from the prior quarter and down 8% from the same quarter prior year due to lower investment yields, which were partially offset by additional investment balances from underwriting cash flow and proceeds from our May 2020 senior debt offering. At quarter end, the investment portfolio duration was approximately 4.7 years, up from 4.6 years in the prior quarter due to both portfolio reallocation and longer duration on recently purchased securities.

Moving now to our loss provision and credit quality. As noted on Slide 14, the mortgage provision for losses for the fourth quarter of 2020 decreased to $56.3 million compared to $87.8 million in the third quarter of 2020 and an increase from $34.4 million in the fourth quarter of 2019.

As shown on Slide 15, we had approximately 15,000 new defaults in the fourth quarter of 2020 compared to approximately 21,000 in the third quarter of 2020 and approximately 11,000 in the fourth quarter of 2019. In addition to the loss provision related to new defaults in the fourth quarter, we reported modest positive reserve development of $7.7 million, with all of this development related to defaults originating prior to 2020. The default-to-claim rate assumption on new defaults remained at 8.5% for the fourth quarter of 2020, unchanged from the third quarter of 2020 and an increase from 7.5% for the fourth quarter of 2019.

As shown on Slide 17, approximately 66% of new defaults in the fourth quarter and approximately 74% of all defaults at year-end were reported to be an a COVID-related forbearance program as of December 31, 2020. We have shared additional information on forbearance program mechanics related to these loans on webcast, Slide 17. These forbearance programs are positive for our industry and for homeowners as they are intended to keep people in their homes through what is expected to be a temporary economic disruption. I'll also note that of our total defaulted loans, approximately 95% of these loans are estimated to have at least 10% homeowners' equity. This factor, along with improving overall economic indicators such as home price appreciation, lower unemployment, governmental support, ongoing forbearance programs and having some end in sight for the COVID-19 environment, help make us cautiously optimistic about the ultimate claim levels. It is important to remember that our reserve estimate is based upon the best available information we have at the time, which includes both external economic metrics and the outcomes of our own proprietary models. As we noted at the beginning of the pandemic, our loss reserve is an estimate of future claim payments, which, under normal circumstances, will not be realized for several years. The broad availability of mortgage forbearance options in 2020 and continuing into 2021, may serve to extend the time line for claim development.

On Slide 15, we show the percentage of new defaults from each of the previous 5 quarters that have tiered as of December 31, 2020. As noted, approximately 61% of defaults from the second quarter 2020 and 50% of defaults from the third quarter 2020 have already cured. Earlier this month, we released an update for January operating statistics that showed a further decline in our primary default inventory as the number of new defaults was relatively flat to December, while cure activity continued to exceed new defaults. Our January cure activity represented 122% of the new defaults reported in the month.

Now turning to expenses. Other operating expenses were $81.6 million in the fourth quarter of 2020 compared to $69.4 million in the third quarter of 2020 and $80.9 million in the fourth quarter of 2019. The increase in operating expenses in the fourth quarter of 2020 compared to the third quarter of 2020 was primarily related to a $6.5 million increase in nonoperating items and adjustments to share-based incentive compensation expense.

Moving now to taxes. Our overall effective tax rate for the fourth quarter of 2020 was 17.9%. The decrease in our effective tax rate for the fourth quarter was primarily due to the effect of a onetime discrete item recorded following the successful completion of an IRS exam of our 2016 and 2017 tax years. Our annualized effective tax rate for 2021 before discrete items remains generally consistent with a statutory rate of 21%.

Now moving to capital and available liquidity. As previously announced, in October 2020, Radian Guaranty entered into a fully collateralized reinsurance agreement with Eagle Re 2020-2 Limited. This reinsurance agreement provides for up to $390.3 million of aggregate excess of loss reinsurance coverage with initial risk in force of $13 billion recovered policies that were issued between October 1, 2019, and July 31, 2020. As of the end of the fourth quarter of 2020, Radian Guaranty had PMIERs available assets of approximately $4.7 billion and our minimum required assets were approximately $3.4 billion. The excess available assets over minimum required assets of $1.3 billion represents a 40% PMIERs cushion. We have also noted on Slide 20, our PMIERs excess available resources on a consolidated basis of $2.7 billion, which when fully utilized represents 80% of our minimum required assets as of December 31, 2020.

The reduction in minimum required assets attributable to the 0.3 multiplier on the required asset amount factor, which reduces the minimum required assets on applicable COVID-19-related delinquencies by 70%, was approximately $650 million at year-end. This has contributed to the significant PMIERs cushion at Radian Guaranty as of December 31, 2020, and we expect that the application of this multiplier will continue to materially reduce Radian Guaranty's minimum required assets for COVID-19 defaulted loans. The expected future benefit to Radian Guaranty, however, is expected to continue to diminish over time. As a reminder, this benefit has thus far peaked in the second quarter of 2020, when we received an approximate $1 billion reduction in the minimum required assets. Our current period PMIERs cushion, including this benefit, is 40%, and excluding this benefit, our PMIERs cushion would be approximately 17%. As of December 31, 2020, we have reduced Radian Guaranty PMIERs minimum required asset requirements by $1.4 billion by distributing risk through both insurance-linked notes reinsurance and other third-party reinsurance arrangements, as noted on press release Exhibit L.

For Radian Group, as of December 31, 2020, we maintained $1.1 billion of available liquidity. Total liquidity, which includes the company's $267.5 million credit facility, was $1.4 billion as of December 31, 2020. It is important to note that most of the cash flows of the parent company are funded by long-established regulator approved expense, interest and act sharing agreements with its subsidiaries and not through dividends from subsidiaries. This provides us with an enhanced level of certainty and predictability in parent company cash flows.

Radian remains committed to managing excess capital in a responsible manner in light of the economic landscape. We have a strong history of taking thoughtful, prudent and shareholder-friendly actions in managing our sources and uses of capital. We have continued to pay a dividend to common shareholders throughout the pandemic, returning approximately $100 million to shareholders over the past year.

During 2020, we returned approximately 25% of our net earnings in dividends to our shareholders. We continue to be encouraged by the trends we are seeing in the business today and believe we are well positioned to leverage the strength of our existing overall capital position. As a reminder, we returned $226 million to shareholders via our buyback program prior to temporarily suspending the program on March 19 due to the pandemic. And we still have purchase authority of up to $198.9 million available under the existing program, which does not expire until August 31, 2021. Given the capital strength that Radian Guaranty and the financial flexibility provided by our available liquidity at Radian Group, we believe that we are well positioned to grow our businesses and deliver value to our shareholders. I will now turn the call back over to Rick.

Thank you, Frank. Before we open the call to your questions, let me remind you that our success in 2020 is a result of our outstanding team that was able to not only meet the challenges of the pandemic environment personally and professionally, but also successfully support our customers and each other in a demanding high-volume market. We focus on supporting our communities through thoughtful and high-impact charitable contribution programs in 2020, both as a company and in support of our employees, we will continue to do the same this year. We grew our valuable insurance portfolio to $246 million and increased book value per share by 11%. We've fortified our capital position and increased our financial flexibility to return $326 million of stockholders through a combination of dividends and share repurchase. Finally, our business model weathered the storm and demonstrating that through the cycle, resiliency we and the mortgage industry have been building since the last financial crisis.

Now operator, we'd like to take questions.

(Operator Instructions) And our first question comes from Randy Binner from B. Riley.

I had a question about the commentary around new insurance written. Rick, I think you sized the MI market at $450 billion to $500 billion in NIW. I just wanted to clarify that I had that right? And if there was also kind of a guide to how your NIW might change year-over-year?

Yes. Thank you, Randy. I appreciate the question. And yes, that is the size we see in the market given pretty varied forecast across the GSEs and the Mortgage Bankers Association. I think -- so as we kind of look at the market year-over-year, we do see it declining, right? Overall originations declining. We see that driven by refinances declining, which ultimately is a decline that will improve our persistency as well. And I think we see a strong purchase market going to next year, continue to be fueled by first time homebuyers, requiring low down payments. So we continue to see a strong mortgage insurance market, fueled by a purchase market, probably smaller than next year. And we've not given any guidance relative to our own kind of forecast for NIW. But I think, look, we're going to participate in the market, and we're going to get our fair share of the market based upon where we see economic value. And I think -- again, I think it's going to be one of the -- if it's going to be probably the second largest -- potentially the second largest NIW market ever. So I think next year continues to look strong.

Okay. And my follow-up is on the refi piece and understanding the MBA data, but I guess -- there's a 2-part question. One, I mean, there was an expectation of lower refi going into this year. I'd be curious what you've seen so far in '21 relative to initial expectation? And related to that, I mean, is it -- are you seeing lenders out there who have an agenda to grow possibly kind of pushing refi more than we'd normally expect. Meaning all in, is it possible that the refi is going to exceed expectation this year above MBA levels?

Yes. I think the MBA versus some of the other forecasts are largely driven by interest rate paths, right, in terms of how they view the progressing interest rates through 2021. And so I think that's going to largely drive it. There also is a factor, the spread, the primary, the secondary market spread could tighten, providing the opportunity for continued refinances. So we're watching all of that. And I think rates still are relatively low, even though we've seen them back up a little bit. And I think it's too early to tell the path of refinances. I think from our -- our viewpoint is that purchase mortgages are more likely to have MI. So we're more focused on the purchase market. Certainly, refinances drive persistency. And I think our expectation is that, yes, refinances continue, but ultimately over the year, probably slow down, which will drive improved persistency overall across our industry. But it's very hard to forecast and predict. I think, we are in a point in time today where there are so many macroeconomic factors that play interest rates. And obviously, the COVID environment in -- just there's a shortage of housing supply, right? So I think you could see the purchase market would be a lot stronger if it was able to meet the demand. But there are just many factors. So I'd say refinances are going to probably be with us, and we would expect them to slow through the year, but I'm not here to predict the direction of interest rates. I'll let others do that.

Our next question comes from Bose George from KBW.

First wanted to just ask about your insurance in force growth. It's moderated even with the number you noted, it's about 4% versus an industry, which grew around 8%, It looks like our NIW share has shrunk a little bit as well. Just curious, do you think some of this is being driven by not being active in the bulk market? Any color on share would help? And also, can you just talk about some positive trends in the market?

Yes, thank you. And Derek and I will tag team this one. But I think as you think about insurance in force of our business, yes, there is some -- I'll let Derek comment on the market share kind of splits because I do think some of that does drive us to where we choose to compete. And we don't think all NIW is created equal from an economic value point of view, but again, I'll let Derek come back to that. But overall, our insurance in force year-over-year grew 2% and if you take out some of the servicer reconciliations, I think it was around 4%. And I think as you think about our insurance in force portfolio, we came a couple of years ago, I think our single premium policy mix is probably 30 or higher. I don't have the exact number. We're down to about 22% today. And so what we've seen is really the transition and repositioning of our insurance in force to a much greater percentage of monthly book. So we probably have a disproportionate share of the single market coming into these refinances, which -- so we -- the expectation that we would see that portfolio kind of churn at a higher rate, it's consistent with our thoughts. We've seen a mix fall. But I think the other thing that's interesting is that it's -- we've been able to, as a single prepaid. So we saw our monthly business grow 11% from a portfolio perspective. We saw our single premium business decline 21% year-over-year, but in that decline, we're able to recognize the earned premium on an accelerated basis. We're also able to return in free capital for investment in the other parts of our business. So we've seen that transition from a higher percentage of singles to, let's say, a more consistent level in monthlies happen over 2020. In fact, there's an interesting fact. Frank and I both mentioned 40% of our insurance in force is relative to the 2020 book. But if you look at our monthly book, 45% of our monthly book is from the 2020 vintage. And so we're seeing this transition as pivot into a more monthly focused book, which we think is good for the long term. And going forward, I think you're going to see not only the '20 book kind of play an increasing role in our insurance in force. But as we head into 2021 with another strong purchase market, you're seeing really the reshaping of our insurance in force towards something that going forward given the macro market economic kind of fundamentals that we see, we believe the continued expansion of a low purchase or low down payment purchase market providing a great backdrop for the market going forward. We see this pivot as really kind of evolving our book to a strong monthly premium business, high-quality, low coupon that we expect to -- has significant economic value, and will continue to provide strong earnings profile going forward. So we -- I would just characterize our insurance in force growth is somewhat of a pivot and a repositioning towards a monthly book, which I think bodes well for the future. So I'll let Derek comment on the market share.

Sure. Thanks, Rick. In terms of market share, one of the things you alluded to were the bulk bids, and that can affect market share. So to the extent that more of the origination market went to those originators where we're not participating, that can have an impact. The volatility around market share is not unexpected. We talked about that, especially as the industry moved towards a black box pricing framework. You can expect some quarter-to-quarter volatility. But long term, what we expect to see in the industry is kind of a distribution more towards kind of a pro rata distribution with some volatility. This quarter, not surprising, you looked at it kind of pre-COVID pricing was pretty stable and then COVID hit, there was a lot of adjustment in terms of pricing. So kind of across the industry, pricing was adjusted upward. And in terms of our black box pricing in Q4, we still had pricing elevated, probably, in aggregate, more than 10% relative to pre-COVID levels. We think some of our competitors kind of gave back more of their price increase probably in Q4. And again, that's not unexpected as the macroeconomic environment becomes more positive, you'd expect to see kind of that pricing adjust. And so when we're looking at our focus is really, as Rick had indicated, a long-term economic value. So what we're always trying to do, kind of expecting a long-term kind of pro rata distribution. We're looking for that share of the market with the highest relative value. And when we're trying to find that, we're looking at not only our view in terms of the ultimate performance from a return on capital perspective, but also very mindful of where the market clearing pricing levels are. What price do we need to set to that to win that business? How does that translate from a return perspective? And so any time you're trying to find those spots, we're constantly adjusting our price up and down to kind of find those spots. And quarter-to-quarter, sometimes your market share will go up when you're kind of finding those spots, sometimes it go down quarter-to-quarter. The main thing is it's a large market, and we find the business to be accretive. So we're just going to continue to kind of execute on our strategy. We think that's the best strategy and most accretive from kind of a long-term shareholder value perspective.

Okay. Great. Definitely very helpful. And then just a quick other one, just on the financial services segment. Just the run rate from this quarter, how should we think about the run rate from that segment and that other line item just as we model '21?

Yes. I think you're talking about real estate services?

Yes. The mortgage services segment, yes.

Yes. Yes. So there's really 2 components. I'll take the real estate side just real quickly. I think, look, we continue to invest in these businesses. And I think we're at a point today where we're going to continue to do that, and I'm happy to talk about why. But I think what we saw in the fourth quarter was really -- really for the full year there was some impact from the pandemic environment around our REO business and really our valuation business, specifically around BPOs and as driven by the eviction and foreclosure moratorium that have been in place throughout the year. Those are businesses that have strong revenues and margins. And they're really poised for a comeback as those moratoriums are lifted because we have kind of a strong market position in both of those. I'd say our title business we're focused on investing and scaling that business. So when we look at the growth in that business that we're beginning to see in terms of our -- many of our MI customers coming over to the title clients. And as we look to scale that business, we are staffing ahead and investing in technology. And so that's going to continue as we scale that business, Bose. And I would say you've seen us carve away the Clayton business and kind of our traditional AMC business, but we are highly, highly focused on our digital valuation and real estate businesses, and we're investing in advanced technology across those businesses to really leverage the assets that we have. And so I'd say, look, we'll provide more detail as we get through the year, and we expect to provide more information on our Investor Day, which we'll announce for later this year.

But I think the business is today, we're going to continue to invest towards growth and value. And there's going to be a little bit of volatility. These businesses are not going to move the EPS needle in the short term. But as I've said before, we do believe that they are building value ahead of kind of ultimate financial contribution. And I think a good indication of the businesses that we're in, is all you have to do is kind of look at what's going on in the fintech and proptech world today from a valuation point of view. And our business is squarely fit in that digital mortgage and real estate ecosystem. Today, they align very well with our mortgage insurance business because today, we get data from them. They're accretive to our customer relationships. We have great market insights across the mortgage and real estate business, what's happening, you can see in the Radiant home price index. But today, that's a value we get while we're investing towards being part of that whole transformation towards the digital mortgage and real estate ecosystem. And we think these businesses have embedded value. And I think the market observations around fintech, proptech businesses, I think we're well positioned from a shareholder value perspective to continue to invest and develop those business towards future value. So from a going-forward point of view, we're going to continue to make investments. I'm not going to manage it on a quarter-to-quarter EBITDA point. We're making investments in these business towards our our view of how best to create value from a shareholder perspective, and we'll continue to do that. But at our Investor Day, when we get -- when we announce the date for that, we'll provide much more detail on this, that would be our plan.

And our next question comes from Mark DeVries from Barclays.

Frank, I think I ask you about capital returns every quarter, so I'm sure you're well prepared for this. But I just wanted to hear your latest thoughts on your appetite to resume buybacks here just given the favorable trends we're seeing in the defaults and also the substantial liquidity at the holding company as well as the large PMIERs excess at Radian Guaranty.

Thanks, Mark. And yes, I appreciate the consistency of your question. So -- and look, we're going to respond similar to I have in the past, which is our thoughts on capital planning, really have remained consistent over time. And that's to balance our organic growth, make sure we've got an adequate risk buffer, make sure that we're considering any strategic opportunities to make sure our debt to total capital level is appropriate. And I think what you've seen over the last 6 years and even through the pandemic, we have been able to address those priorities and still return capital to shareholders, and our dividend has been uninterrupted. But we did pause our share repurchase program, as you noted, at the height of the pandemic when there was elevated uncertainty regarding the broad economy and potential losses. But I will say now that we do have greater visibility toward the future and have seen improvement in the economic landscape over the past several quarters. The overall economic environment does seem to support a return to normal, if you will, approach to capital management. And so in that regard, as a reminder, and to the extent that we might reengage in any repurchase activity. Our philosophy would remain a value-based approach regarding the repurchase pricing levels. And so we do have the flexibility to initiate a new 10b5-1 plan at our discretion in the future during an open trading window. And we do have remaining authorization of $198.9 million that doesn't expire until August 31. So yes, I would say, overall, things are looking better, overall. And so I hope that's helpful.

Yes. That's very helpful, Frank. And then just a question on the yield, the premium yield, really helpful color and also disclosure on Slide 11. Just was hoping, though, Frank, you could give us a little bit more color on what we should expect kind of directionally and maybe even dimensionalize some of the moves we could see in the different lines, the in force portfolio yield, the single premium cancellations and the ceded earned premium lines to help us think about where kind of the net might move under different assumptions. I mean, I clearly understand, like, for example, the premium cancellation has to do with expectations around prepayments and runoff of the singles. But anything you could provide to help us think about kind of all those different moving pieces?

Sure. I think each of those components is -- has different drivers, if you will. I think we've continued to guide this really for many quarters now, that we do expect the overall portfolio yield to continue to decline. And when I -- we won't give guidance on the magnitude of what that decline would be. But I think you've noted some of the drivers there, the single premium cancellations certainly is a big driver there. The ceded premiums is another. So I think those -- you would need to have a view on those. But as Rick said, less of our portfolio is single premium now. So -- and the opportunity for refinance activity is something you have to put your view to predict that. So I wouldn't want to put too fine of a point on anything other than just to say directionally where we've continued to guide. And of course, all of that, especially on the in-force yield on the pricing on the NIW and the yield that we're getting on new business is dependent upon mix as is what's coming on. So it's difficult to predict, but again, I hope that's helpful in some way.

Yes. Well, if I could just follow-up on that. If we just take like the ceded earned premium line, I assume that the drag there is still being impacted by higher losses and the profit commission. Should we think about that as things normalize, moving back closer to that 4.5 points we saw back kind of pre-COVID. Is that the right way to think about; that one?

Yes, I think that's right. As we do return to what normal looks like, I think many of these things would continue to be on a relative basis, similar.

And our next next question comes from Mihir Bhatia from Bank of America.

Maybe I'll start with Slide 15 on the cure rate. I just wanted to confirm that we're looking at this information correctly. Just taking Q2 as an example, the cure rate data suggests. So I think it's 60% of 2Q delinquencies have cured. I believe your initial default-to-claim ratio on those was 8.5%, right? So is the expectation in your reserving that about another 30% will cure? And then, I guess, relatedly, can you just talk about how that 60% number compares to your -- what you had been expecting at this time, maybe even like last quarter, like as you have been seeing the trends develop?

Sure. This is Frank. I think your interpretation of the slide is correct, that 61% for the 2Q '20 [61.1%] have cured. I think the thing that's challenging is -- and we've said this really since the beginning of the pandemic, and we saw the surge in defaults, is we've got an estimate of what the ultimate claim rate will be. Where we don't have clarity into is the timing of when that will occur. So I wouldn't want to be too precise about any expectations there. But I think given a little more time, some of that might become a little more predictable just based on what our experience actually is. But -- and that's why we thought it would be helpful to provide that information on that slide.

And I think -- to -- Mihir, to your question, I think it's -- that run rate towards the 8.5% ultimate decline rate or claim rate is kind of what we're working towards. I don't know. I think we saw this morning a kind of a quick news flash about the FHFA also extending forbearance potentially for another 3 months out to 18 months. So some of those -- people have an incentive to continue to kind of stay in that forbearance state for an extended period of time, to the extent they can. So I think back to Frank's point, it's just the timing part of it is something that we're all just -- we're monitoring and watching. But I think programs are generally positive for the market, allowing homeowners to get back on their feet and kind of get through this economic stress caused by the pandemic. So I think we generally continue to feel good about the progress we're seeing.

Great. No, that is actually quite helpful. And then just a couple of quick ones. The Texas -- the Texas event last week, fair to assume that, that will be a non-event from an MI perspective, right?

Yes. We don't see that as a material little bit in terms of the overall portfolio.

Right. Got it. Okay. And then just last question on persistency. I know it kind of depends on interest rates. And you've talked a little bit about the size of the market being lower because refis go down. But any numbers you would be willing to put on where you think persistency gets to towards the end of 2021, maybe at a minimum, at least, do you think it's troughed for now?

Yes, this is Frank. It really is hard to say on what near-term persistency is going to be. I think we continue to say that longer-term persistence in the low 80s is still what we would expect. But near term, I think the wildcard is certainly refinance activity. So wouldn't want to try to get too precise on the near-term.

And our next question comes from Ryan Gilbert from BTIG.

I also saw that FHFA press release, it looks like they extended the forbearance period to 18 months and extended the foreclosure moratorium to June 30. And I'm just wondering if you had any kind of initial comments on how that would impact your business and how we should think about the impact to loss reserving, in particular, the prior year reserve in 2021, just thinking about all the reserving that's been done in 2020? And then also on capital return potential in 2021 if the extension of the moratorium or the forbearance period would impact your capital return decision-making process?

Sure. This is Frank. I'll start, and Rick and Derek can add in on the forbearance extensions. But it is, as we've said before, the forbearance programs are very good for us. It keeps people in their homes. It prevents a claim payment from occurring. So anything that occurs to keep people in their homes, especially over longer periods of time to let more home price appreciation develop, et cetera. All of that bodes very well for us. So those are all very positive things for the industry overall. So we feel good about that. As it relates to reserving, our reserves, as I mentioned in the prepared comments, are made, it's our best estimate at a moment in time. It takes these things into account. But I would say we feel obviously very good about where our reserve levels are now. They're appropriate for the risk that we see. And those things certainly can change over time as new facts and circumstances and trends developed. So I would say things are certainly trending in a good direction, how that ultimately makes its way into what our estimates on the ultimate claims, we'll just have to wait and see how those play out. And then I think your last question was on capital returns. And Mark had asked about this previously. So I think we feel good about our current capital levels, our holding company resources and certainly, the trends that we're seeing broadly speaking. So I think we feel -- we feel like many things are starting to reach a normal. And I think the history would show that we're very good stewards of capital and returning capital to shareholders when it's appropriate. So Rick or Derek, I don't know if you want to add anything on the forbearance.

Sure. I was just going to add, I mean, in terms of the extension on the forbearance period and then the foreclosure moratorium, not surprising, we think there's a good recognition among policymakers that kind of time is helpful. So not only do you have the home price appreciation from our perspective, but just in terms of the vaccine rollout, reemployment rate. So giving borrowers that are having temporary difficulties more time to kind of write that is a positive. Also, in terms of just the options to exit forbearance are also very borrower and MI friendly, and we continue to see those exits to be quite positive to date. So overall, as expected, and I think a good recognition in the industry as to how best to deal with the kind of current situation.

Okay. Great. I really appreciate that. I did just want to circle back on the capital return question because I'm just wondering if the foreclosure moratorium extension or the forbearance extension influences your decision-making on capital just around the timing of capital return, not necessarily the absolute level of capital itself. But when do you think you could increase capital return to shareholders?

Yes, I don't know that it would necessarily influence timing. I mean, it's just another positive and a list of positives that we're looking at. So all of these things are taken into account in totality.

Yes. I would just add, Ryan, when you think about this, the relief program going through you think about all the support in the marketplace putting politics aside. These are all good from a macroeconomic point of view in terms of supporting the homeowners and our small businesses and communities to get back on their feet. I think that just adds to the general positive trends in the macroeconomic scenario that we're all starting to see. But we still have to be cautious because there's still uncertainty in the market, but I'd say the net-net these types of changes that are very supportive to providing kind of a transition for people in harms way are going to ultimately be positive from a macroeconomic point of view.

Okay. Got it. And then second question was just around the January new default trends. I think you mentioned this in the prepared remarks, but I was a little surprised to see it kind of flat lined from December. And I'm wondering if we should expect further improvement in the months ahead? Or if you're seeing further improvement in February or this level of kind of new defaults is the new post-COVID normal?

Yes. This is Derek. So in terms of that, no, you did see some month-to-month movement and also you have some seasonal effect because we're adjusting pretty quickly, as Frank had indicated, in terms of the new default they're down dramatically from Q2, almost 80%, they are down. So they're getting closer and closer to kind of what they were pre-COVID, but they're still about 30% above. We would expect kind of through the year, assuming continued kind of positive economic trends, assuming that there's not somehow a turn downward, we would expect to see that kind of positive development on a year-over-year basis, understanding that there's going to be your kind of typical seasonal effects because we are feeling good from a macroeconomic perspective. The housing market is still extremely strong recovery. Embedded equity is very strong in terms of the portfolio. You have unemployment rate going down. And so all of those kind of macro factors continue to point in a positive direction and how it would impact new defaults.

And our next question comes from Geoffrey Dunn from Dowling & Partners.

First question. Obviously, 2020 had some unique considerations from expenses with respect to travel, et cetera. So Frank, can you, given that consideration, maybe help frame up the expectation for MI expenses, operating expenses in '21?

Sure, Geoff. So expenses [is,] I think, if you recall, last quarter, and this was after the sale of Clayton, we estimated at that time, quarterly expenses sort of a normalized run rate basis would be about $70 million a quarter. And I think if you look back through 2020, you'll see that that's mountain average out, actually, it was less than that, about $68 million a quarter. So I would say, going forward, a similar type of run rate, is it $70 million plus or minus $1 million or $2 million, that's probably the right level. The thing that is always difficult to predict, as you know, there are some variability due to usually compensation-related items. And then also, we do recognize an expense benefit from the ceding commissions under our single CSR. So periods of high singles cancellation can elevate that benefit. But generally speaking, that's where I would put you -- subject to some change, but if there is an expectation of a run rate, which we'd certainly bring that to your attention.

Okay. And then with respect to your MI reserving methodology, my sense is, I'm not sure if you ever talked about it directly, but that the industry was very conservative with the home price depreciation assumption, at least back in Q2, and I think it was probably flat to down. Has your HPA assumption and your reserve assumptions evolved since Q2?

And roughly, if you can share kind of what are your assumptions going into the reserves in Q4 and as you enter '21?

Yes, Jeff, it's hard to pin anything on a particular assumption. I would say that we look at all of these assumptions in totality and make our assessment accordingly. And so you saw no change in the default-to-claim rate period-over-period. And I think that's consistent with our view. As things change, we'll adapt as we need to. But for the moment, that's what we felt was most appropriate for the period.

I understand the claim rate. I'm talking more on the severity assumption and how you're factoring an HPA into that?

Yes. Derek, if you've got a thought on there.

Yes. Geoff, it's Derek. No. So in terms of severity, we kept that pretty stable quarter-to-quarter. And obviously, the HPA, it does express itself significantly in the default-to-claim rate as well. Often, it's more of a frequency than a severity. We do think over time, though, given kind of the home price accretion embedded equity that could have an impact, but we haven't realized any of that in our reserve assumption by decreasing the severity and taking that into account to date.

And our next question comes from Jack Micenko from SIG.

Rick, earlier, you talked about picking your spots in a big market. And look like on the NIW more recently, some of the FICO score has migrated down a bit, but maybe LTVs have gotten stronger in terms of the mix of the business. Is that the right interpretation in terms of what the pricing model is kind of putting out? Is LTV becoming a bigger part of the consideration in the model than FICO in terms of predictability of future loss? Just curious if I'm reading too much into the press release numbers or if there's a more pronounced shift underway?

Yes. No, I think -- appreciate the question, Jack. I think -- and Derek can pick up on this and fill in on the details. But I think, as we've mentioned, we focus on where we see economics across our customer base and across our distribution. So I think you're going to see month-to-month, quarter-to-quarter kind of changes based upon our pursuit of economic value. But let me just pass over to Derek, and he can talk about things she season progressing.

Yes, Jack, I mean, one of the things I talked about earlier was you kind of have to looking on two dimensions. One dimension is what we think kind of the modeling as our ultimate performance. So those our prediction in terms of default probability and prepayment probability, right? So that's not really changing so much in the model and what would move it kind of the bigger factor oftentimes is just the competitive landscape, as I referred to before, what's the market clearing level. And at times, that market clearing level was going to shift around the credit spectrum. Sometimes it might be in higher FICO buckets and lower FICO buckets, meaning market clearing levels that we find attractive from a return perspective. So generally, if you're going to kind of see movements like that, which I don't think that we really have a significant change in terms of any of those dimensions, it would be more so probably driven by competitive dynamics and where we want to allocate the capital going back to where we find the highest relative value. That's much more likely as to what you're going to see when you see those change in trends.

Okay. And then maybe asking Geoff's question a little bit differently. It seems like we're coming out of forbearance at a better clip than we thought. But at some point, there's going to be this sort of terminal number of forbearance DQs that -- and I think we're all waiting to see what happens when they switch the lights on at the end of the night, and who is really a default who goes back to paying. If we kind of progress through '21 and we get another 3 months and another 3 months and no one's going to lose a house under a Biden's administration. Like does the loss reserving, can it pivot more towards HPA? Because Frank, you talked about like a 90 LTV kind of number, I think, or less. Do you anticipate moving more towards an HPA weighted approach, if we just don't get to the end of the line on what these forbearance DQs look like in '21?

Yes. I'll start, and Derek can add in. I think it's -- I think it is tough to play that out, Jack. I mean, I think what you're saying, in theory, makes sense, but there are obviously some more moving pieces there. I think what you're describing is a claim just can't happen and what does that look like. And I -- we'll -- I think we have...

The actual prices keep going up high single, low double, yes. More than -- just like there's just more embedded equity there for that model.

Yes. Look, Jack, that's exactly right. And if you look at our kind of defaulted population right now, I'd say, around 95% probably have embedded equity of 10% and around 65% have embedded equity of 20%. And you're right, that continues to increase. Now you always -- and that's unusual in a recession. So sometimes when you're trying to model that out, right, that can be a little bit tricky because usually, home prices are going down. But to your point, yes, that is certainly a positive trend with respect to it, and that would certainly kind of all things being equal, have downward pressure in terms of that ultimate roll rate. So when you look at it, and I also wouldn't frame it as an either/or, right? So when you're kind of looking at roll rates, you're taking into account multiple things, so not only HPA, things like reemployment rates is going to matter with respect to that. And just some of the realized trends because you always have a certain amount of model there. So you're kind of looking at historical trends, you're trying to predict it going forward, utilizing things like embedded equity, future equity, reemployment rates in terms of those who are actually in default. And then also just looking at the policy landscape. So I would say it's not like a single dimensional thing. So it's kind of -- I wouldn't think of it that way, hey, we're going to switch necessarily. We might weigh certain factors relatively more or less. And I would agree. Home price appreciation, one would think would have more of an impact in terms of your ultimate roll rate in these circumstances than has historically been the case.

And back on Derek's point earlier about affecting frequency because ultimately, borrowers have more ways to exit their property today in the strong house market. Putting equity incentive in it. So I think these are all factors that go in as we think about ultimately default claim rates. And also remember, the longer this forbearance period is extended, the claims still come years after that, right? So it's -- we're out multiple periods before there's actually a cash claim paid, which is another interesting factor in all those. So a lot of pieces to the puzzle, but Jack, I think we're going to continue to watch it, continue to monitor it very carefully, and we'll express our opinions as we go each quarter.

So we have no further questions at this time. I will now turn it over to Rick Thornberry for final remarks.

Well, we appreciate everybody joining the call, and thank you for your thoughtful questions. And we look forward to talking to each of you soon as we have the opportunity, most of all stay healthy and well, and wish you all much success in the coming year, and we'll talk soon. Thank you again. Take care. Bye-bye.

Thank you, ladies and gentlemen. This concludes today's conference. Thank you for participating. You may now disconnect.