Radian Group Inc (RDN) 2016 Q3 法說會逐字稿

Ladies and gentlemen, thank you for standing by and welcome to Radian's third quarter 2016 earnings call. (Operator Instructions)

As a reminder, today's conference call is being recorded.

I would now like to turn the conference over to Emily Riley, Senior Vice President of Investor Relations and Corporate Communications. Please go ahead.

Thank you, and welcome to Radian's third quarter 2016 conference call. Our press release, which contains Radian's financial results for the quarter, was issued earlier this morning and is posted to the "Investors" section of our website, at www.radian.biz.

This press release includes certain non-GAAP measures which will be discussed during today's call. A complete description of these measures and their reconciliation to GAAP may be found in press release exhibits "F" and "G" and on the "Investors" section of our website.

During today's call, you will hear from S.A. Ibrahim, Radian's Chief Executive Officer, and Frank Hall, Chief Financial Officer. Also on hand for the Q&A portion of the call are Teresa Bryce Bazemore, President of Radian Guaranty; Derek Brummer, Executive Vice President and Chief Risk Officer of Radian Group; and Cathy Jackson, Corporate Controller.

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 2015 Form 10-K and subsequent reports filed with the SEC. These are also available on our website.

Now, I would like to turn the call over to S.A.

Thank you, Emily. Thank you all for joining us and for your interest in Radian. Before we discuss our third quarter results and the trends impacting our businesses, I'd like to highlight a notable accomplishment for our Company.

I'm pleased to report that we wrote $15.7 billion of new mortgage insurance business in this quarter and set a Company record for the highest volume of flow MI in our Company's nearly 40-year history. I'd like to thank our customers for their continued support and loyalty, as well as congratulate our entire team for our success in achieving this milestone for our Company.

The high quality business written in the third quarter is expected to generate attractive returns and extend our future earnings momentum. In addition to strengthening our MI franchise, this business creates the foundation from which we are able to develop new growth opportunities outside traditional mortgage insurance.

These growth opportunities leverage our core expertise in credit risk management, our unique insight into mortgage and real estate trends, and our extensive customer relationships. This is the result of our Clayton-Radian combination and is an important differentiator for us among our MI peers. We'll talk more about the opportunities we are seeking outside of our traditional businesses shortly.

Turning to the quarter's results, earlier today we reported net income for the third quarter of 2016 of $83 million, or $0.37 per diluted share. Adjusted pretax operating income was $140 million for the third quarter of 2016, and adjusted diluted net operating income per share was $0.41.

Importantly, book value per share grew for the fifth consecutive quarter, to $13.47, an increase of 14% over last year.

And now, turning to the mortgage insurance segment. We continued to improve the quality of our large mortgage insurance in-force book, which drives future earnings for Radian. As I mentioned earlier, we wrote $15.7 billion in new mortgage insurance business in the third quarter, an increase of 40% compared to the $11.2 billion written in the third quarter of last year, and we set a record in the quarter for our highest volume of flow MI. This record volume adds to our existing post-downturn, in-force book which is among the largest in our industry.

The size of the MI market expanded again in the third quarter, with an estimated total of approximately $75 billion to $80 billion. Based on this market growth, our record third quarter volume, and our significant new business pipeline, we now expect NIW in 2016 to exceed $48 billion, a significant increase over the $41 billion written last year.

We remain focused on writing as much high-quality new business as possible while maintaining a well balanced portfolio mix that we expect to generate a mid-teen levered return on required capital.

We grew our mortgage insurance in force by 4% over the third quarter of last year, and we continue to expect our in-force book to grow and enhance our strong foundation for future earnings.

Radian's success in writing a large share of high-quality and profitable new business after 2008 has helped to improve the composition of our mortgage insurance portfolio and has served as a differentiator within our industry among legacy peers. For the third quarter, our total primary insurance risk in force consisted of 87% of business written after 2008, or 79% excluding HARP volume. You may find these details on webcast slide 10.

Slides 11 and 12 compare the loan characteristics of Radian's existing risk in force as of the year-end 2003, 2007, and 2011, with risk in force at the end of the third quarter of 2016. These comparisons are important, given that our risk in force composition reflects the environment in which the loan was underwritten. The high-quality loans written during the more than seven years following the downturn reflect a much tighter credit environment and more stringent underwriting requirements as compared to the relatively weaker credit environment and underwriting standards experienced pre-downturn.

Turning to opportunities in the market for new volume, as I mentioned earlier industry sources have again increased their market projections for 2016 and are now expecting mortgage originations of up to $2 trillion. In 2017, while the mortgage market is expected to be smaller than it was in 2016, economists are now predicting the largest purchase origination market in 10 years, with purchase originations projected to exceed $1 trillion.

This is good news for our industry, given that private mortgage insurance is three to four times more likely to be used for purchases than for refinancings. Based on this growth in home purchase activity but expected decline in overall mortgage origination volume, we are expecting a modestly smaller total MI market in 2017 compared to 2016.

Turning to business opportunities beyond traditional MI, we continue to focus on ways to leverage our core expertise in credit risk management to write more business and strengthen our MI franchise. In September, we chose to participate in a new Freddie Mac credit risk transfer pilot program which is designed to provide additional coverage beyond traditional levels and to transfer risk to the MI companies. Earlier this month, we agreed to participate in a similar program with Fannie Mae.

Our current commitment under these pilot programs will result in premiums and required capital that are immaterial. Importantly, though, we are confident that the combined risk analytics and business intelligence gleaned from our Radian and Clayton combination helps us as we seek to take advantage of new opportunities and expand our business beyond traditional mortgage insurance.

That said, these programs do not represent the loan-level, front-end, deep-cover MI that our industry has been advocating for that is currently being considered by the FHFA. Our trade association, USMI, recently responded to the FHFA's request for input on front-end credit risk transfer, which we feel would strongly complement the programs already available in the market today. These loan-level, front-end programs offer substantial benefits for taxpayers, lenders, and homebuyers.

Turning to the credit environment, Radian's total number of primary delinquent loans declined again, with a year-over-year decrease of 18%, as you can see on slide 17, and our primary default rate fell to 3.3%.

Turning to our mortgage and real estate services segment, we reported third quarter 2016 total revenues of $44 million, which is an increase both over last quarter and over the same quarter last year. Over all, the segment had a strong quarter, with solid activity in single-family rental and increased volume and margins in loan review and due diligence services.

What's most important to remember is that the services segment adds a diversified source of fee-based revenue for Radian, and the Clayton family of companies broadens our participation in the residential mortgage value chain with services that complement our MI business.

We continued to see progress in our cross-sell initiatives for Radian and Clayton. Cross-sell revenues totaled $2 million in the third quarter and represented a significant increase over last quarter and last year.

Importantly, at this week's Mortgage Bankers Association annual conference, a record 1,400 guests attended the Radian-Clayton reception, and many customers personally expressed their appreciation and regard for our sales and service teams. Customers continue to view our Clayton offerings as a positive differentiator and, in some cases, as the reason to begin using Radian as their MI provider.

And finally, turning to the recent M&A activity and broader industry landscape, we believe the new interest in our mortgage insurance industry in recent months validates the strength and opportunity that we've always seen in the MI model and the prospects for our future.

And while there are no recent updates on the progress of housing reform legislation, there is an increasing focus on the future of the GSEs and on reducing taxpayer risk through credit risk transfer programs. Several members of Congress continue to emphasize this approach. This month, the period closed for comment on the FHFA's request for information on various credit risk transfer structures.

Our industry has received significant support, in particular for our ability to provide loan-level deeper cover which would increase coverage to 50% of a loan's value, versus the standard coverage today of approximately one-third. The Mortgage Bankers Association, Urban Institute, National Association of Home Builders, Community Mortgage Lenders of America, Credit Union National Association, and the Housing Policy Council all support loan-level deeper cover. We look forward to hearing more about the FHFA's analysis of next steps.

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

Thank you, S.A., and good morning. Before I begin, let me first reiterate S.A.'s comments that we have had a great quarter, and I look forward to sharing some of the details that support our continued optimism for our business.

Secondly, let me draw your attention to press release Exhibit D, which was added in the second quarter and is intended to provide further visibility into certain items impacting our premiums earned and other operating expenses. This quarter, for other operating expense detail, Schedule D presents the total expense for a particular line item, not just the variable component as we presented in the prior quarter.

So, let's begin our discussion with the key drivers of our revenue. New insurance written was $15.7 billion during the quarter, compared to $12.9 billion last quarter and $11.2 billion in the third quarter of 2015.

As S.A. mentioned, our third quarter was a Company record for flow NIW, and the new business we are writing today has excellent credit characteristics. For example, more than 64% of NIW in the third quarter consisted of loans with FICO scores greater than or equal to 740, and only 5% with FICO scores below 680.

As S.A. also mentioned, we expect our level of this high-quality new insurance written for the full year to exceed $48 billion.

Direct single-premium business represented 27% of our total NIW in the third quarter, which was essentially flat to the second quarter. Our retained single-premium exposure this quarter was 17%, net of the insurance ceded with our singles-only quota share reinsurance transaction.

Refinancings increased to 22% of volume this quarter, compared to 18% last quarter, and were up from 13% a year ago.

Due in large part to the increase in refinancings this quarter, our 12-month persistency decreased from 79.9% in the second quarter of 2016 to 78.4% in the third quarter of 2016, as noted on Exhibit L. With respect to how persistency is currently trending, our annualized three-month persistency decreased from 78% in the second quarter of 2016 to 75.3% in the third quarter of 2016.

Our long-term normalized persistency expectation of the low 80%s remains unchanged, but given the recent refinance activities due to continued low interest rates, we may not see that level return for several periods.

Primary insurance in force increased to $181.2 billion during the quarter, a 4% increase over the same period last year.

Net earned premiums for the quarter increased to $238.1 million in the third quarter of 2016, from $229.1 million in the second quarter of 2016. This increase was the result of three factors: the accelerated recognition of unearned premiums on single-premium policy cancellations, of $3.6 million; approximately $1 million in related profit commissions under the single-premium QSR; and our growth in insurance in force.

It is also noteworthy to mention that net premiums written for the third quarter were $241 million, compared to $232 million in the second quarter of 2016.

Investment income remained relatively flat in the quarter and was impacted only slightly for actions we took for the anticipated use of the $211 million in cash needed to redeem the 2017 senior notes.

Total services revenue for our mortgage and real estate services segment was up over comparable quarters, at approximately $44 million for the third quarter 2016 as compared to $39 million in the second quarter 2016 and $43 million in the third quarter of last year.

As we've mentioned previously and as you can see on slide 20, we expect fluctuations in revenues in this business as the transactional and seasonal nature of these businesses contributes to greater volatility in revenue, though we have begun to see an increase in volume for the single-family rental securitization market that positively impacted our real estate valuation and component services business. In addition, we experienced encouraging revenue growth from loan review and due diligence activity, as well as margin improvement in this business.

Moving now to our loss provision and credit quality. We remain optimistic about the trends we're seeing in our credit quality. As such, it is important to note that our primary risk in force consists of only 13% legacy business originated before 2009 and that those vintages are contributing positively to earnings, as you can see on slide 13.

This portfolio composition is unique, and it can skew overall performance metrics where a legacy versus non-legacy analysis may be more informative. For example, we continue to see the majority of our new defaults coming from our legacy portfolio, which represented 73% of the new defaults in the third quarter. While total new defaults declined 2% year over year, new defaults from our legacy book fell 13%.

It is important to note that only 14.5% of defaults in our total portfolio are from loans written after 2008. And as these newer books of business written after 2008 are beginning to reach peak default years, we are seeing an expected increase in default activity from those books, though at very low loss levels, as you can see on slide 14.

We continue to see improvement in credit trends and fundamentals in our overall portfolio. For example, our cure rate has continued to improve on a year-over-year basis. For the third quarter of 2016, our cure rate was 16.7%. So far this year, we have seen the highest cure rates in the past seven years. This included a 6.1% cure rate on our defaulted loans that are more than 12 payments past due, the second quarter since the downturn that this rate has exceeded 6%. This also included a post-crisis high of a 2.3% cure rate on pending claims, which represents claims submitted that were subsequently withdrawn.

And our default rate fell to just 3.3% in the third quarter, which represents a decline of 20% year over year.

Given this continuation of positive trends observed during the quarter, our estimated default-to-claim rate on new defaults was reduced to 12%, as compared to 12.5% in the prior quarter. As our legacy portfolio continues to shrink, changes to our default-to-claim rate apply to a smaller and smaller universe, thus driving a relatively small dollar change when these assumption changes are made. During the quarter, several other small assumption changes were also made that largely offset the approximate $6 million positive development attributed to the default-to-claim rate change.

On slide 15, you'll find the breakdown of the components of our loss provision. This separates the impact of losses on default notices received in the current quarter from the impact of reserve development on default notices received in prior periods. Generally, the impact of the provision from current-quarter defaults is driven by new defaults, as disclosed in our roll forward, multiplied both by our estimated default-to-claim roll rate for new defaults and by our expected average claim paid.

Claims paid for the full year 2016 are still expected to decline significantly from recent years, to approximately $375 million.

Over all, the performance of our portfolio remains very good, with positive trends continuing, further evidence of a strong credit profile of both post-crisis business as well as greater predictability around the legacy portfolio.

Now, turning to expenses. Our reported other operating expenses for the quarter were $64.9 million, as compared to $65.7 million in the second quarter of 2016 and $65.1 million in the third quarter of 2015. Other operating expenses in the third quarter on a comparative basis to prior quarters were impacted by items that are detailed on Exhibit D.

Incentive compensation expenses were $12.8 million this quarter, down from $14.2 million in the second quarter of 2016. These expenses, as presented on Exhibit D, represent total expenses associated with both our short- and long-term incentive programs but do not include direct volume-related incentives such as commissions.

Incentive compensation can be volatile from quarter to quarter. For short-term incentives, management makes periodic accrual adjustments throughout the year based on our estimates of full-year performance relative to targets established at the beginning of the year. Additionally, as we saw in the second quarter of this year, incentive compensation expenses can be elevated due to grants for our equity-settled, long-term incentive awards and the associated required acceleration of the related expense for retirement-eligible employees. Exhibit D incentive compensation totals include approximately $6 million in total for both of these noted items during the third quarter.

Ceding commissions on reinsurance transactions in the third quarter of 2016 were $5.5 million, compared to $5 million in the second quarter of 2016 and $1.3 million in the third quarter of 2015.

We maintain our expectations that fourth quarter 2016 operating expenses will reflect the successful application of our targeted efficiency initiatives that we've disclosed previously.

We do, however, anticipate our variable operating expenses to be proportionally higher, given our higher-than-expected mortgage insurance business volumes. This further highlights the scalability of our business model and represents a positive for our shareholders. These volume-related variable expenses for the third quarter of 2016 were approximately $3 million higher than in the same quarter of 2015.

Over all, we are pleased with the progress we've made and the discipline we have exercised in managing our expenses to our expected 3% to 5% reduction we had targeted, which had assumed NIW levels that were largely consistent with the prior year. At Radian, we have a culture of continuous process improvement which serves our customers, our shareholders, and our employees very well.

Moving now to taxes. We currently estimate our annualized effective tax rate for the year before discrete items to be approximately 38%, above the statutory rate of 35% primarily because [of the] material portion of the losses related to our convertible debt repurchases that are non-deductible for tax purposes.

Our actual effective tax rate for the third quarter was 34.8%. The overall reduction in our third quarter effective tax rate was primarily driven by our 2015 return to provision accrual adjustment, which is normally recorded as a discrete item in the third quarter of each year.

And lastly, moving to capital, our recently announced capital activities and those completed this quarter are further indications of the improved outlook for both our Company and our industry. We have demonstrated discipline and prudence to date and we'll continue to be opportunistic and deliberate in our management of capital.

We have also largely completed the capital plan that we outlined last year whereby we sought to improve our capital structure by removing the convertible notes and distributing our debt maturities more evenly over time.

We continue to move forward on our path to returning to investment grade at the holding company. On September 28, 2016, we received a ratings upgrade from S&P for both Radian Group and Radian Guaranty, putting us one step closer to our goal for Radian Group. Radian Guaranty continues to be an investment-grade rated company by both Moody's and S&P.

Radian Guaranty's PMIERS available asset cushion is approximately 7% of minimum required assets, consistent with our expectations. Keep in mind that in addition to the cushion at the operating company, there is also accessible capital at the parent company that currently represents another 6% of minimum required assets after our expected future capital actions, based on our projections and the current requirements under the PMIERS. It is also expected that Radian Guaranty will continue to build available assets organically each quarter.

On June 29, Radian Group announced a plan to execute a $125 million common equity share repurchase program. Subsequent to that announcement and during the third quarter, the Company adopted a Rule 10b5-1 plan to implement the program.

Due to the timing of the implementation and the parameters of the 10b5-1 plan, Radian did not repurchase any shares of its common stock during the third quarter. As the parameters of the plan are value based, the pace of future repurchases is uncertain.

This authorization is expected to remain effective through June 2017, and Radian has the ability to recalibrate the parameters of the 10b5-1 plan periodically, though changes to the plan can only take effect during an open trading window.

Additionally, we purchased approximately $21 million of face value of our outstanding 2.25% convertible senior notes due 2019 and settled those in cash, reducing our diluted share count by approximately 2 million shares.

As we announced last quarter, we also redeemed the remaining $196 million aggregate principal amount outstanding of our 9% senior notes due 2017. Radian Group paid an aggregate redemption amount of $211 million, including accrued interest through the redemption date.

The combination of all of our 2016 capital actions has decreased the Company's total number of diluted shares by 23.1 million shares, or by approximately 10%.

After all of our planned uses of cash, our available holding company liquidity at the end of the third quarter would be approximately $250 million, though keep in mind that the timing related to some of these actions may extend over the next 12 months.

As a result of these actions, Radian's debt-to-capital ratio is now well below our targeted 30%, at approximately 27%.

Our plans for future capital actions will be in the context of positioning the Company for future growth while being mindful of further rating agency upgrades and shareholder preferences. We expect further significant improvement in our capital and liquidity position over time given the health of our mortgage insurance business, and we will keep investors well informed of our plans.

I will now turn the call back over to S.A.

Thank you, Frank. Before we open the call to your questions, let me remind you that we set a record for Radian in writing the highest volume of new flow MI business. This new business helps fuel our insurance in force growth and extends our earnings momentum.

We benefited in the quarter from positive credit trends, including another decline in our total number of delinquent loans, high cure rates, and outstanding performance from our newer books of MI business.

We successfully completed capital actions in the quarter and the year that have improved our capital structure and collectively decreased our total number of diluted shares by nearly 10%.

With that, Cynthia, we would like to open the call to questions.

(Operator Instructions) Eric Beardsley, Goldman Sachs.

Just a question on the claim rate. You mentioned taking it down from 12.5% to 12%. Just wanted to get your thoughts, I guess, in what you needed to see over the next year, or so, or a few quarters to bring that down further and ultimately where you expect that to settle out?

Sure. So, at the beginning of the year I think we had guided to an expectation of roughly 100 basis points of decrease in the default-to-claim rate, which we have fully experienced this year. I think as we continue to look at that, if we see continuation and improvement in the trends that we're seeing thus far, we have reevaluated every quarter. But again, as those trends would continue, we would expect to see that reflected in the assumptions in the model.

Got it. And then, just on the benefit you've been having from the single-premium policies prepaying, I guess if you were to stay in the, call it, high 70%s persistency rate, where should we expect that number to be? And I guess, at what point does it become much less significant?

I think if you turn to Exhibit D, we've actually called out that item separately. And I think you can match up what the acceleration is relative to what persistency levels have been and just make your own best estimate of what that is. Certainly, as we continue to write a mix of business between monthlies and singles, naturally we'll see prepayments, I think, for some time, as we've indicated.

So, there is what I would call a normalized level, though I wouldn't hazard a guess at what that is. But certainly in the period, we saw an elevation.

Got it. So, if we were to have the MBA forecast play out next year, with a 40% decline, or so, in refi, do you have any rough estimate on where that could go?

I don't.

Okay. Great. Thank you.

Doug Harter, Credit Suisse.

Sort of along the same lines, but with rates having backed up this quarter, how quickly does persistency move and follow rates?

Again, that is really hard to estimate. I think we have seen what I would call little mini-refi periods quarter to quarter, and you've seen what the impact has been on persistency. Again, a very tough number to estimate, which is why we're trying to guide more on the longer term, over time, low 80%s type persistency rate. But on a quarter-to-quarter basis, it really is a tough number to estimate.

Got it. And then, switching to the share buyback, anything you can give us as far as the expected pace at which that $125 million will be used?

And that number, too, is a difficult one to estimate. Our 10b5-1 program is a value-based plan, and it is tough to estimate, certainly, what a future stock price would be, how the pace of repurchases would perform across a range of value scenarios there. So, very difficult to estimate.

All right. Thank you.

Mark DeVries, Barclays.

First, just to follow up on that last point, Frank, if the stock price just held at current levels, do you have any sense for what pace the repurchases would be done?

We don't provide any color or commentary around the specifics of the parameters in the 10b5-1 plan. So, I'd want to stay away from that question, specifically.

But I would just remind you with our earlier repurchase program we were able to accomplish that one in a very short time horizon, because the stock was trading very low relative to book value and tangible book value. We bought back shares under the last repurchase program at about 88% of book value and right at tangible book value. But that was, I would say, a unique set of circumstances during the last one.

But again, I don't think I can provide much color beyond what we've said on the 10b5-1 plan.

Okay. Do you have any flexibility to purchase outside of that plan, as well?

We do not. But as I said in the prepared comments, we do have the ability to adjust the parameters. But those changes can only take effect during an open trading window.

Okay. And so, just to clarify one of your other points, I think you indicated you've got about $250 million of liquidity at the holding company after intended uses. Does that include both the $125 million on the repurchases and also buying back what remains outstanding of the converts?

It does, yes. So, absent those items we have roughly $485 million in cash.

Yes. Okay. And if the stock price holds at these levels for another 10 days, should we expect that you're going to call the remaining 2019 converts?

We're certainly mindful of the call feature of those converts, and we're paying attention to it.

Okay. Fair enough. And then, for that remaining $250 million of liquidity, how should we think about that? I know you've been hesitant to kind of indicate what a normal buffer is, but should we think the rest of that would be available for additional share repurchases once you've gone through the current authorization?

No. What we've said previously -- and this is in conjunction with our capital plan and our quest to return to investment grade -- we do expect to maintain some constant level of parent company liquidity. I think previously we've said between $300 million and $350 million.

But in the context of the PMIERS cushion, in particular, we expect to continue to build that organically at the writing company. And the additional holding company liquidity that we have, technically it's available as an additional buffer. And on a combined basis, we've said on a combined between Radian Guaranty and Radian Group if that buffer is between 10% and 15%, we're comfortable there.

But, no, I would set the expectation that we would expect to maintain a certain level of cash on an ongoing basis at the parent company.

Okay. But that's why you're building the buffer within Radian Guaranty, right? So, at some point, if you reach that buffer inside Radian Guaranty you wouldn't necessarily need to hold as big of a cushion at the HoldCo?

Not necessarily. I would say the capital planning for Radian Guaranty at some point diverges from the capital planning at Radian Group. And again, technically, the Radian Group liquidity is available for the cushion, but as that cushion builds at Radian Guaranty we wouldn't necessarily expect to free up the cash that we're holding at Radian Group.

Okay. Got it. All right. Thanks.

And let me just further clarify. The reason for maintaining that $300 million to $350 million at Radian Group is to have an amount that is sufficient to repurchase any single year debt maturity, and that's important as we think about rating agency [consideration].

Bose George, KBW.

Actually, I didn't know if you mentioned this in the text, but do you have an estimate for your market share? It looks like you might have taken some share, at least relative to the peers that have announced already.

Sure. Teresa?

We actually don't know that number yet, until we see the reporting from everyone come through. So, while we're excited about the amount of business that we were able to write over the quarter and we would hope that we've seen an elevated market share, we won't know that until we get all that data.

Okay. And presumably, in terms of any fallout from the Arch-United Guaranty merger, that's also to be seen what that does to your market share?

That's absolutely right. Arch said that they expected to see some reduction in the overall combined market share. And as you can expect, we're constantly focused on gaining share from a day-to-day perspective. So, we would certainly be looking to get as much of that, sort of, disrupted share as possible.

We continue to focus on our differentiation where we believe we've got strong relationships and service and the benefit of the Clayton products, and so forth.

But we haven't really seen that yet, and that's not atypical given the fact that that merger hasn't occurred yet.

Okay. That makes sense. And then, actually, switching to the average premium, I wanted to -- what's the average premium on the new insurance written currently, just given that your credit mix keeps improving?

Sure. So, on the portfolio basis, I think we've said previously that we expect to have that hover right around the 50 basis point level. But on the new business, it really does depend upon the mix of new business that we're writing. So, some quarters it may be a little below the 50 basis points; some quarters it may be a little above.

But I think the important point to remember here is that our rate card is calibrated such that irrespective of the FICO-LTV bucket and what the basis point charge actually is, the returns are more evenly distributed across the rate card. So, on a return basis, there really is no change quarter to quarter in what our expected returns are on new business.

Okay. That's great. Actually, just one more, on the mortgage services segment. The improvement in the revenue, is that related to your earlier comment about the improvement in single-family securitization activity? And then, just sort of going forward, how should we think about when that gets to better profitability? Is that from revenues? Is that from expenses? Or, just more color on that would be great.

Sure. So, the single-family rental securitization certainly was a contributor to the volume, and we had some other businesses there pick up as well, as I shared in my prepared remarks.

And I think what we do expect to see as that revenue increases, margins should expand as we scale up from that level of fixed costs. The variable costs associated with it are relatively low. So, we should see margin expansion there.

But, yes, I would call it just basic blocking and tackling that will get us to improve performance on the near term. And then, certainly, if there is a return to private-label market, that certainly would be a significant change in the profile of what drives the revenue over all. But I wouldn't expect that to happen in the near term.

Okay. Great. Thanks.

Phil Stefano, Deutsche Bank.

I just wanted to revisit a conversation a couple of callers back about the building of capital at Radian Guaranty versus Radian Group. And maybe you can help us think about, are the conversations starting to happen about potentially upstreaming dividends? Are those not important conversations to have at this point? Maybe just some thoughts around that as the capital cushion continues to build?

Sure. It's certainly important as we think about our overall capital planning for the next several years. The cushion will continue to build in Radian Guaranty.

And I think we've discussed previously that at some point we would expect to approach Pennsylvania with a request for a special dividend. But at this time, given where our holding company liquidity is, given the other creative sources and uses of capital that we've had with the surplus notes, et cetera, we don't see that happening in the foreseeable future. But we have a very good relationship with our Pennsylvania regulator and would expect, if the need arises, to present a compelling case to request a special dividend. But we don't see that happening in the near future.

Phil, also keep in mind that our holding company liquidity needs on a day-to-day basis are minimal, because we have an expense-sharing agreement with our subsidiaries. So, we do not need to get dividends to pay for the holding company expenses on a regular, routine basis.

Okay. And that dovetails pretty nicely into the next question I had looking at the allocation to corporate expenses. It feels like 2016 has been a bit more heavily weighted toward services. If I recall correctly, first quarter 2016 there was some commentary about severance charges and that might be skewing it. But it feels like there's been an uptick in management's time spent looking at the Clayton business. Is that being driven by maybe some of the things we're seeing in headlines? Or, is there anything else that's kind of caused a more permanent shift in management's time at services?

I would not describe it as anything situationally specific. So, periodically, we go through a time allocation exercise [within group], and we make an estimation of the time that's spent in different places of the business to help guide us in our allocations. I wouldn't read too much into that number, but that is an accounting estimation that we use to facilitate our segment reporting.

And at a high level, as I commented earlier, we are looking at leveraging the combined capabilities of our mortgage insurance (inaudible) risk insights with the broader [vision] we have through the (inaudible) businesses in developing new and leveraging new ways of delivering risk management services to our customer, and more about that in the future. We're always looking at creating future opportunities.

That's a great point. Thanks, and best of luck.

Vik Agrawal, Wells Fargo.

I think you said that you're expecting 2017 NIW to be slightly lower than 2016, and I think you mentioned roughly a $1 billion purchase market. If we see rates as we're seeing now back up, if we see another 25 basis points back-up in mortgage rates, how do you kind of view that, given the 2017 expectations you laid out?

I'll ask Teresa to answer the question on any volume-related things, but in my comments I did not mention NIW volume. What I said was we expect the mortgage origination market to be modestly lower next -- the MI portion of the mortgage origination market to be modestly lower, with two moving parts: the refi business shrinking dramatically next year -- that's what the projections show; but the purchase market being the highest in the last 10 years, which is a very positive for us.

And the net effect of the two, I mentioned, would be a modest reduction in the size of the market available to MI.

Now, Teresa?

I think the only thing that I would add is that you also talked about the fact that as MIs we do three to four times better in a purchase market. And so, that's why we would expect to only see a smaller effect on the MI market than the overall market might see. So, that's really what we commented on.

But we are not projecting NIW for next year at this point.

Okay. Thanks for your clarification. And then, on our Q3 NIW, sort of the LTV composition shifted a little bit. Is that more indicative of what you're planning to see in the future? Or, is that just sort of a one-time anomaly here?

I think that in some respects we were expecting to see a little bit of uptick in terms of the [97s], if that's what you're talking about. There's been a lot of focus on sort of expanding credit and having programs that would allow for those programs. So, I don't know that we would see large upticks in that in the future, but I think we were all expecting to see some uptick given the focus there.

Okay. Thanks for the comments.

Chris Gamaitoni, Autonomous Research.

Congratulations on the strong NIW quarter. I was wondering if you could give us some color where really the performance came from? Was it a couple of large customers, bank channel? I ask that as the purchase growth was similar to other players that have also reported, but refi was certainly outperforming its part. So, I'm trying to understand what exactly happened in your origination channel that drove that outperformance?

So, I think there were a number of different dynamics. So, it's sort of hard to say kind of all of them.

But I would say that part of it is just that we had said that once the pricing in the industry sort of leveled out, that we would expect to see Radian gain in share, because we felt like we could differentiate ourselves based on our relationships and our service, and so forth. And so, I think we've seen that play out now in both the second and third quarter once that happened.

We continue to see growth in the credit union segment; that's something that we talked about focusing on.

And also, we have very strong relationships as well with some of the large non-bank originators who have continued to show growth.

So, I think it's coming in a number of different ways, but that's exactly what we were talking about when we said we thought that we could differentiate ourselves with price being kind of on a relatively level playing field.

Perfect. And then just a follow-up, maybe for Derek. I was wondering if you could give us any color into the reasons you're seeing for the 2.3% cure rate on pending claims? Is it the theory of documents are unable to be found to prove the claim is valid? Or, what's the change in trend you're seeing on that level?

I think in part it actually has to do with home price appreciation. And so, oftentimes, what you'll have is in those cures claims submitted, but the lender might find if they actually keep the property and don't submit a claim, it might be more advantageous just to keep the property and dispose of it. So, I think that's some of what you're seeing in terms of that uptick.

Makes sense. Thank you so much.

Geoffrey Dunn, Dowling & Partners.

Teresa, if we believe the MBA forecast for originations, it looks like we might have seen an uptick this year in purchase penetration, just in general for the industry. Is that something you think the industry is experiencing? And what's driving that?

I think we've seen maybe a little bit there, and we've seen it be a little bit higher than we would have necessarily expected. And certainly, we're hoping that that continues.

And was that more first-time homebuyers coming into the market? Or, what's shifting?

That's really hard to tell, Geoff. I think some of it certainly is first time homebuyers coming into the market.

Okay. With respect to FHA, we're a month or so, maybe a little less, out from the annual report. Who knows what happens with the rates, but is there any chatter in Washington about the standards to which FHA's capital is held? Maybe evolving that from the pre-crisis standards?

There's continued to be discussion about that. I think, actually, Dave Stevens at the MBA made some comments on it not too long ago. Because you remember there's that 2% statutory cushion, but in fact, there's a lot of discussion about whether the cushion should be higher than that.

But there's just sort of been dialogue about it. I don't know that anybody is actually planning to do anything specific about that.

Okay. And then, Frank, do I remember correctly that the quota runs through the end of this year and then the new cedes cease?

No, the quota share [recessions] also covers the new business in next year, as well.

Okay. As you're thinking about reinsurance planning then for 2018, obviously there's a leverage benefit to your returns of maintaining reinsurance, but there are alternatives in the market. Any initial expectations for how you might treat the book beyond 2017?

This is Frank. What we said historically -- and our actions have supported this -- is that we will use reinsurance more as a risk management tool and not a capital management tool. So, to the extent that we see development in the portfolio that is getting maybe a little heavily concentrated in one area or another, that is our primary and intended use there.

Okay. Great. Thank you.

Jack Micenko, SIG.

Frank, in your prepared comments you said that the fourth quarter operating expense line will reflect the initiatives that have been taking place. I think you had about $3 million-$3.5 million-$3.6 million of between severance and [tech] and then kind of a $12 million number year to date. Does that imply that those one-timers are going to be burning off here pretty sharply, going forward? Or, how do we think about those other items?

So, on the comparative basis and really thinking ahead to the fourth quarter of 2016, what we've said is that we're really managing the expenses associated with running the business down to that 3% to 5% lower level. So, we'll continue to highlight the things that we consider to be lumpy or perhaps have some timing attributes that are a little unusual.

But I would expect just business as usual will always have some lumpiness to it, and certainly the volume related for this period and for next period, as well. And there are still some expected costs associated with fully implementing our expense plan.

So, hope that's helpful. And it's part of the reason why we've developed Schedule D in the earnings release, just to try to highlight some of those material moving parts.

Okay. But then I guess at a high level, is it safe to assume that these types of expenses are the driver and that over time they should be lower, not higher?

It really depends on the particular item. So, for instance, severance cost, as a good example, is probably elevated this year relative to what a business as usual expectation would be on an ongoing basis. Expenses associated with, again, implementing our expense reduction plan is another type of expense that we would expect to burn off over time. But other costs, again some of the accounting associated with the incentive plans, is just naturally lumpy, and we'll continue to call those out.

Okay. Fair. That answers the question. And then, Teresa, is there any way a front-end plan works without an FHFA [fee] reduction? What should we think about there?

So, I think if you don't -- you could certainly do a front-end without the fee reduction. I don't know how it would compete ultimately from an FHA point of view.

But when you think about it and part of the point here is to reduce the actual risk that the GSEs are taking on in the first place. And it seems very easy to do that when you're talking about risk we already have and just expanding that risk some.

Certainly, our view is that if we're taking on more of that risk, then the GSEs should be charging less in terms of LLPAs, because they have a reduced and more remote risk than what they have today. But I think that it could work, but it wouldn't be as advantageous.

Okay. Okay. And then, one more for Frank. Sorry. What's the path you see to the investment grade at rating group? Are there any yardsticks in your mind you're looking at and saying: Hey, these are pivotal points over the next 12 to 18 months that we should be thinking about.

For the most part, we have achieved most of the hurdles that existed previously. Certainly, the debt-to-total capitalization, getting that below 30%; maintaining some level of holding company liquidity, et cetera. I think from this point forward it really is just the passage of time and an overall industry view that will probably drive most of the future upgrades.

Okay. That answers it. Thank you.

Mihir Bhatia, Bank of America.

Most of my questions have been answered, but if I could just -- maybe if you guys could just spend a few minutes discussing the competitive environment, particularly around LP MI? I think that would be helpful for us.

Sure. So, there's always been competition when it's come to LP pricing. And I think that hasn't changed, although I do believe it's moderated some over the last few months.

So, one of the things I think is probably worth pointing out is that for those of you who may not be familiar California also recently issued guidance to the industry, essentially stating that any discounts to pricing, whether they're lender paid or borrower paid, have to be provided to all similarly situated lenders. So, in other words, you can't selectively discount on a lender-by-lender basis, which we've seen.

And with respect to the impact on us, we recently completed an underwriting exam in California, and California approved the process that we will go through to evaluate pricing eligibility of lenders. So, we don't foresee any problems for us, but we can't speak for others in the industry. But we do believe it could affect some of the others in the industry based on what we've seen in the market.

In fact, let me add from a Radian perspective that [we are] pleased with this development.

Right. So, I guess along those lines then, in terms of the industry consolidation -- and maybe it's just a slightly different topic -- can you discuss not just the market share implications, but do you guys have a view on in terms of what that does towards maybe putting in, bringing in a little more discipline around pricing? And maybe in the different channels, too, because they're going to lose market share? Is there a channel that you expect to gain a little bit more market share in versus the others? I think that would be helpful.

Well, just very briefly, I think if you think about the Arch-UG combination, both of them are using the black box pricing, which has less transparency, and our expectation would be that we would see kind of one black box, going forward. We think that's a positive.

Most of the customers we have like the sort of rate card, risk-based pricing that we have.

But over all, I think that it's hard to sort of say what any sort of changes would mean. We're just continuing to focus on what our customers are looking for and how we deliver the right kind of experience to them.

One of our biggest strengths, which was the reason why when I was a Radian customer I did a lot of business with Radian, is the fact that we have great customer relationships. And you saw the example of that this quarter.

We also, as we have been talking about periodically, have a very focused effort on credit unions.

And then, the other thing that Teresa has talked about in the past is we have a group of inside sales team, which is very unique, that focuses on calling not only smaller customers which otherwise economically we couldn't handle with a traditional sales force, but we also use that to call the loan officers in larger customers.

Got it. Okay. Thank you.

Mackenzie Aron, Zelman & Associates.

I guess just one question, on the risk sharing. Obviously, very small at this point, but can you just maybe talk about the returns on that business, just how it compares maybe order of magnitude to what you would see on the primary MI? And just going forward, assuming that opportunity does grow, how you would think about allocating the capital between the primary MI and potential risk-sharing opportunities?

Sure. So, the first part of that question, our interest in participating in a pilot program was really just to stay close to it and see how the programs develop.

As far as what our long-term participation would be, it really is in the context of what types of returns, what is the risk-return profile, et cetera. So, I think it's premature to talk about allocations of capital there, because our primary focus really is just around generating high returns on our capital.

That makes sense. Teresa, maybe you're the right person to answer this. But just as you continue to kind of face the pushback that the GSEs have had around counterparty, where do you think the biggest opportunity is for the industry to kind of win them over? Is it cash collateral that's the biggest impediment? Or, kind of what do you see as, going forward, the opportunity there to allow risk sharing to become more front end?

Mackenzie, I think we're focused on really making the case to FHFA about why this is the right way to do it, why there is more transparency for the system over all. And while there is a difference of opinion on whether or not it would ultimately translate to a benefit to the borrower, we believe that it's the best avenue for identifying an opportunity to get a benefit to the borrower. So, we're really focused in our conversations, not just with the GSEs but also with the FHFA, to make those points.

Okay. Great. Thank you.

Speakers, I'd like to turn it back over to you for any closing comments.

Thank you, Cynthia, and thank you all for participating in our call. And thank you for the terrific questions and for the congratulations we got. See you next quarter.

Thank you. Ladies and gentlemen, that does conclude your conference call for today. Thank you for your participation and for using AT&T Executive TeleConference Service. You may now disconnect.