Oct 23, 2018
Executives
David Urban – Director-Investor Relations Sandro DiNello – President and Chief Executive Officer Jim Ciroli – Chief Financial Officer Lee Smith – Chief Operating Officer Drew Ottaway – President of Banking
Analysts
Kevin Barker – Piper Jaffray Bose George – KBW Chris Gamaitoni – Compass Point Henry Coffey – Wedbush Scott Siefers – Sandler O’Neill and Partners Daniel Tamayo – Raymond James Scott Beury – Boenning and Scattergood
Operator
Good day, and welcome to the Flagstar Bank Third Quarter 2018 Earnings Call. Today’s conference is being recorded.
At this time, I would like to turn the conference over to David Urban, Director of Investor Relations. Please go ahead, sir.
David Urban
Thank you, Jonathan, and good morning. Welcome to the Flagstar third quarter 2018 earnings call.
Before we begin, I would like to mention that our third quarter earnings release and presentation are available on our website at flagstar.com. I would also like to remind you that any forward-looking statements made during today’s call are subject to risks and uncertainty.
Factors that could materially change our current forward-looking assumptions are described on Slide 2 of today’s presentation, in our press release and in our 2017 Form 10-K and subsequent reports on file with the SEC. We are also discussing GAAP and non-GAAP financial measures, which are described in our earnings release and in the presentation we made available for this earnings call.
You should refer to these documents as part of this call. With that, I’d like to now turn the call over to Sandro DiNello, our President and Chief Executive Officer.
Sandro DiNello
Thank you, Dave and thank you everyone for joining us today. In addition to Dave, I’m joined this morning by Jim Ciroli, our Chief Financial Officer; Lee Smith, our Chief Operating Officer; Kristy Fercho, our President of Mortgage; Drew Ottaway, our President of Banking; and Steve Figliuolo, our Chief Risk Officer.
I’m going to start the call by providing a high-level view of our performance for the quarter, and then I’ll turn to call over to Jim for details on our financial results. Lee will follow with a review of our business segments and strategic initiatives, and I’ll conclude with guidance for the four quarter before opening up the line for questions.
In the third quarter, we once again turned out solid earnings, which demonstrated strength of our banking business. We posted adjusted net income of $0.85 per share in the third quarter, the same as the second quarter and well ahead of the $0.70 per share we made the same quarter a year ago.
Net interest income showed outstanding growth, climbing 8% to $124 million, bolstered by higher earning assets and the expanded net interest margin. Additionally, quarter-by-quarter, we keep building our servicing portfolio, which along with our banking business, provides a stable source of income to offset the volatility of the mortgage business.
We now service approximately 620,000 accounts and expect to exceed 800,000 accounts by year-end. Expenses were another positive in the quarter, declining 2% despite strong growth in the balance sheet and higher mortgage originations.
We did have costs related to our pending acquisition of 52 branches from Wells Fargo, but cost discipline, lower mortgage cost and a better mix of businesses with better efficiency ratios helped clamp down expenses. We also saw actions we took early this year in response to expected softness in the mortgage market fully took shape.
Our mortgage revenues declined $16 million in the quarter, mostly due to a decline in our overall gain on sale margin. This was partially offset by a stronger return on mortgage servicing rates.
While we also suffered secondary margin compression, but channel margin decline resulted in large part because of a shift in mix from retail to bulk production. This change in mix allowed us to reduce expenses effectively thus minimizing the effects of the very competitive mortgage environment.
Given the overcapacity in the mortgage industry and the uncertainty as to when this will normalize we have taken a cautious and thoughtful approach to the business. This has allowed us to maintain our market share, while at the same time, reducing expenses an approach we can take because of growing revenues in our banking and servicing businesses.
Another positive note was that we continue to maintain pristine asset quality. Net charge offs will make $1 million plus the downward trend in loss rates in our held-for-investment portfolio paved the way for us to take a $2 million provision benefit.
We currently have no nonperformance commercial loans as credit quality continues to be superb. Now let’s turn to capital.
The lifting of our Federal Reserve Supervisory Agreement was a major milestone in the quarter. It ushers in a new era for our holding company by providing more flexibility in managing our capital.
Meanwhile, we capitally remain strong. The community bank is prospering, and our diverse business model continues to pay off.
Finally, we look forward to the successful onboarding of the Wells Fargo branch which we expect at the beginning of December, and all the benefits we expect it will bring to our company. Now let me turn it over to Jim.
Jim Ciroli
Thanks, Sandro. Turning to Slide 6.
We’re pleased with where the quarter ended. When you exclude expenses related to our pending acquisition of the Wells Fargo branches, our adjusted net income this quarter was $49 million, $0.85 per share, roughly in line with last quarter’s net income of $50 million and also $0.85 per share.
This quarter’s results were better than last quarter’s performance other than mortgage gain on sale. As we’ll discuss in a couple of slides, we continue to grow the earning assets of the bank and expand the profitability of those assets.
Our earning asset growth was consistent with the growth rate we’ve experienced since the Q4 of 2014. Deposits also grew nicely, and our net interest margin once again expanded and yet we only added marginally to our nonmortgage expense base.
Asset quality remained pristine, yet had an allowance that covered 1.5% of the held-for-investment loans. Mortgage gain on sale revenue came in as lowest we have experienced since the second quarter of 2011.
However, since we have transformed the earnings power of the bank, our net interest income was down nearly 2.5 times what it was then. I’ll delve deeper into this quarter’s gain on sale results from the next slide.
Asset quality continued to be strong as net charge-offs were only 5 basis points. Our nonperforming loan ratio improved yet again.
And our allowance coverage for loan portfolio remained among the strongest in the industry. There continued to be no early warning signals or credit issues in the portfolio.
Capital also remained strong. And we are managing our capital ratios assuming the pending acquisition of the Wells Fargo branches and assuming the capital simplification regulations going to place.
On that basis, we are no longer constrained in our Tier 1 leverage ratio, but rather by our total risk based capital ratio. With a few slides, I’ll take you through a more extensive analysis of that total risk based capital ratio.
So let’s turn to Slide 7 and dive deeper into the income statement. Net interest income benefited nicely from the ongoing progress we have made in growing our banking business.
Earning assets were up 5%, and the net interest margin expanded by 7 basis points to 2.93%, as higher yields on earning assets more than offset higher funding cost. We’ll discuss earning asset growth in the next slide.
We saw the net interest margin expand as deposit growth allowed us to reduce higher cost FHLB borrowings. Also, at the margin, when we can find a funding advantage over comparable term FHLB borrowings, we would use as corporate deposits to bridge our funding needs to December.
And these funds will be replaced with lower cost core funding the branches to be acquired. We experienced a very challenging mortgage market in the third quarter.
Noninterest income decreased 13% as mortgage revenues fell $16 million. Our gain on sale revenue fell $20 million, while the returns on mortgage servicing assets improved by $4 million.
The decrease in gain on sale revenues primarily reflected lower GOS margin as it fell to 51 basis points and dropped 20 basis points, primarily due to secondary marketing performance during the quarter. The margin also weakened from a mix shift towards lower margin, but lower cost delegated to both correspondent volume and fewer RMBS deals in this quarter compared to the prior quarter.
The percentage of purchase mortgage volume at 71%, was consistent with prior quarter. The net return on the mortgage servicing assets improved $4 million, resulting in a net gain of $13 million this quarter as compared to a net gain of $9 million last quarter.
The increase from last quarter largely reflected lower prepayments, stronger valuations and expenses were well controlled. Excluding $1 million in transaction costs attributable to the pending branch acquisition, noninterest expenses $172 million, a decline of 3% from the prior quarter, reflecting lower mortgage expenses and cost-reduction initiatives.
Commissioned and loan processing expenses were a combined $5 million or 13% lower than in the prior quarter, even these total loan originations were 1% higher in this quarter. The shift in mix towards the lower cost delegated for a correspondent channel and away from the retail and broker channels that are experiencing higher levels of pricing compression also contributed to this expense reduction.
Credit costs were negligible again in this quarter. We experienced only 5 basis points net charge-offs.
Consumer loans that were more than 30 days past delinquent and still accruing totaled only $3 million and nonperforming loans were $2 million lower at $25 million or 28 basis points of the portfolio, while the commercial loan portfolio remained clean. As a result of all these positive metrics, the allowance came in $3 million lower at $134 million.
So let’s now turn to Slide 8, which highlights our average balance sheet this quarter. Average loans held-for-investment rose $492 million or 6%.
We saw broad-based growth in all major categories. The commercial loan portfolio experienced solid gains in warehouse, commercial real estate and C&I loans.
For consumer loans, we continue to portfolio mortgage loan production such as professional mortgage loans that produced nice risk-adjusted returns on equity. Also, our indirect marine/RV portfolio totaled $121 million at the end of the quarter and continued to demonstrate good traction among dealers.
Average deposits rose $922 million or 9% of the quarter, primarily led by higher custodial, brokered and commercial deposits. Average custodial deposits rose $366 million.
The resulted increase in service accounts this year and seasonal factors. On a related basis, average commercial demand deposits were up a solid $259 million as we expanded our commercial deposit relationships primarily from mortgage servicing businesses.
As we pivot to risk based capital ratio constraints, these deposits bring us reductions in the risk weighting and qualified loans to the same borrowers, effectively, A, providing risk based capital and B, providing a funding at a cost that is favorable to margin at alternatives. As mentioned before, we’ve been using brokerage deposits to bridge our funding needs to early December when these funds will be replaced with lower cost, more stable branch deposits.
At September 30, our tangible common equity to assets ratio remained strong at 7.74%, while our tangible book value per share rose to $25.15 as we continue to compound our earnings. We will cover regulatory capital shortly.
So let’s now turn to asset quality on Slide 9. Asset quality continues to be pristine and there are no early warning signals of a change in that trend.
Early-stage delinquencies continued to be negligible. Only $3 million of consumer loans were over 30 days delinquent and still accruing at September 30, unchanged from June 30.
We currently have no commercial loans over 30 days delinquent and still accruing, and no commercial nonperformers. At September 30, our allowance coverage was 1.5% total HFI loans.
Coverage remained strong at 1.6% of consumer loans and 1.4% of commercial loans. This commercial loan coverage is even stronger because 31% of that portfolio is in warehouse loans, which are fully secured and we hold the collateral.
Turning to Slide 10. Capital remained robust.
Our total risk-based capital ratio was 14.2%, an improvement of 16 basis points from last quarter. Net income commercial deposits against qualified related commercial loans benefited this ratio at 31 basis points and offset the cost of balance sheet growth in the period.
We anticipate that this ratio will be at 13.7% at December 31, assuming the closing of the Wells Fargo branch acquisition, the recognition of already realized deferred hedging gains currently in AOCI and enactment of the Capital Simplification regulations. Similarly, our Tier 1 leverage ratio is 8.36% at September 30, down 29 basis points from last quarter, resulting primarily from balance sheet growth and higher MSRs, partially offset by earnings retention.
We anticipate that this ratio will be at 8.8% at December 31, again assuming the closing of the Wells Fargo branch acquisition, the recognition on deferred hedging gains, pending enactment of the Capital Simplification regulations. What these pro forma capital ratios suggest is that we will continue to have a robust level of capital with the lifting of the Supervisory Agreement two months ago, we have more leverage than we can pull to optimize our capital.
Further, I continue to remind you we have a low-level of risking our business, stellar asset quality with strong allowance coverage, a neutral rate risk position and strong liquidity levels, combined with the low cost risk management function. As such, we would expect an upgrade at least in the short term in the lower half of our target range of 13% to 14% in the total risk-based capital ratio.
Also, with the total risk-based capital constrained, it could make sense for us to issue subordinated debt in the near future to support asset growth. I’ll now turn to Lee for more insight into each of our businesses.
Lee Smith
Thanks, Jim, and good morning, everyone. We’re very pleased with our adjusted operating net income of $0.85 per diluted share for the quarter.
We grew average earning assets $793 million or 5% and expanded our net interest margin by 7 basis points, which enabled us to grow net interest income $9 million quarter-over-quarter. Net return on our MSR asset improved $4 million in the quarter, and we again demonstrated our ability to manage expenses in a challenging mortgage environment.
All these meant we were able to keep earnings consistent with the second quarter after adjusting for the integration costs associated with the pending Wells Fargo branch acquisition despite gain on loan sale revenues declining $20 million quarter-over-quarter. I think this result is testament to how we have diversified our business model and made our earnings more predictable, while growing the community bank and our subservicing businesses over the last few years.
In addition to our solid earnings, I also want to update you on some of our key strategic initiatives. We expect to close the Wells Fargo branch acquisition by the end of the day on November 30.
These branches will provide us with over $2 billion of high-quality, low-cost deposits, which effectively pre-funds our balance sheet growth for the next two to three years. We’ve been working diligently since we announced the deal to ensure seamless transition for both customers and employees.
We’re also leveraging our experience from the Desert Community Bank branch acquisition earlier this year where we’ve seen no deposit runoff in the seven months since the deal closed. Average total deposits increased $922 million in the quarter, predominantly a result of higher custodial balances because of the higher subservicing loan count.
Through the first nine months of 2018, we’ve grown average total deposits by $2.3 billion and seen healthy growth across all deposit categories. We’ve been very successful in 2018 creating liquidity to fund our projected balance sheet growth over the next three years.
At the end of the quarter, we were servicing or subservicing approximately 620,000 loans. This represents an increase of 40% since the beginning of the year, and we’re scheduled to onboard another 200,000 non-Flagstar originated loans in the fourth quarter, which will push our loan accounts comfortably over 800,000 by year-end.
Finally, we entered into an agreement with GreenSky during the quarter to originate unsecured home improvement loans through the GreenSky program. This illustrates how we continue to diversify the way we can add loans to the balance sheet.
It’s also an extremely efficient asset class for us given the originating and servicing of the loans is done through the GreenSky program. It’s been another solid quarter as we continue to expand our footprint and grow of our balance sheet with high-quality interest-earning assets.
I’ll now outline some of the key operating metrics from each of our major business segments during the quarter. Please turn to Slide 12.
Operating highlights for the community banking segment include: average commercial loans increased $253 million or 5% in the quarter, and we saw balance growth across all lending channels; average warehouse loans increased $91 million or 6%; average commercial real estate loans increased $89 million or 4%; and average commercial and industrial loans increased $73 million or 6%, continuing our transition for strong community bank. Average consumer loans held-for-investment increased $239 million in the quarter as we added high-quality residential first mortgage loans, nonauto indirect loans, and HELOC’s through our portfolio.
Going forward, we will continue to grow our consumer loan portfolio through the same channels, plus the addition of the GreenSky program home improvement loans I mentioned earlier. Average interest-earning assets have increased $2 billion or 14% over the last 12 months.
And we believe we can continue to grow at the same rate over the next two to three years given we have the necessary liquidity, capital and asset generating capabilities from our unique business model to support this growth. Of this $2 billion in earning asset growth, $1.3 billion has been driven by high commercial loans, with the remaining $700 million being driven by consumer loans held-for-investment and loans available-for-sale.
Net interest income has increased $9 million or 8% quarter-over-quarter and is $21 million or 20% higher the net interest income in Q3 2017. We’ve been pleased with the continued growth in both interest-earning assets and net interest income over the last several quarters.
Average total deposits increased $922 million, or 9% in the quarter, led by an increase in average custodial and commercial deposits. Finally, we’re excited to be closing the Wells Fargo branch acquisition at the beginning of December and we have been working diligently to ensure a smooth transition for our new customers and employees.
As of September 30, there was approximately $2.12 billion of deposits attached to the 52 Wells Fargo bank branches that we are acquiring versus the $2.28 billion we announced in June. The majority of the attrition that has occurred to date has been nondomicile residents, i.e., customers that originally lived in the same geography as these branches that have since moved away.
The remaining attrition is a combination of normal course retail activity and seasonality of balances. While difficult for us to predict overall attrition, we still hope to do best than the 17% of post-closing attrition we previously disclosed.
We only have to pay a premium on the average daily deposit balances during the 30-day period preceding the closing day. So we don’t pay a premium on the attrition that has occurred to date.
Given updates to other assumptions as we have become more informed, we expect similar compelling aspects of the transaction. The deal that is moderately accretive to 2019 earnings per share and has a payback period of less than five years.
We are very encouraged by the performance of the community bank, our ever expanding footprint and potential for continued future growth. Please turn to Slide 13.
Operating highlights for the mortgage origination business include: fallout-adjusted lock volume decreased 8% to $8.3 billion quarter-over-quarter, while the net gain on loan sale margin fell 20 basis points to 61 basis points. As a result, gain on sale revenues fell $20 million to $43 million in Q3 versus $63 million in Q2.
The challenging mortgage environment and excess operating capacity across the industry continues to put pressure on margins. Our net gain on loan sale margin fell 20 basis points quarter-over-quarter to 51 basis points primarily due to secondary market compression, a change due to channel mix as we originated more delegated bulk loans, which have a lower margin for the more efficient from a cost to originate point of view and the competitive mortgage market.
We do believe the problem with excess capacity within the industry will eventually correct itself, and that should help with some of the margin compression we’ve experienced this year. Moving to servicing.
Of course, the operating highlights for the mortgage servicing segment on Slide 14 include: we serviced or subserviced approximately 620,000 loans, of which over 490,000 are subserviced for others, making us the 7th largest subservicer in the country. We’ve already increased the number of loans serviced by 176,000 or 40% so far this year.
And given our existing onboarding schedule for the fourth quarter, we estimate we will be above 800,000 loans serviced or subserviced by the end of this year. So far in 2018, we sold or agreed to sell $28 billion in MSRs, including a pending sale of $5 billion in the third quarter and retain the subservicing on 100% of these sales.
The custodial deposits that these loans generate also help us fund our loan growth. We have $2 billion of such deposits during the quarter, up $366 million or 23% from the prior quarter.
We will continue to grow our subservicing platform through a combination of MSR sales where we subservice the underlying loans and onboarding non-Flagstar originated loans given the strength of our platform and one-stop shop offering. We’re thrilled with the performance of our servicing business so far this year and are very close to increasing our servicing capacity to 2 million loans.
Moving on to expenses on Slide 15. Our noninterest expense decreased $4 million to $173 million quarter-over-quarter, which was primarily driven by a lot of compensation benefits, commissions, and loan processing expenses as we took actions to manage expenses in line with the lower mortgage banking revenues.
Our adjusted efficiency ratio remained relatively flat at 74% quarter-over-quarter. We estimate noninterest expense will be between $185 million and $190 million during the fourth quarter, which includes further one-time integration cost of $10 million as well as one month of run rate operating expenses from the pending Wells Fargo branch acquisition, which will close at the beginning of December.
As always, we remain focused on improving our efficiency ratio through increasing revenues, while maintaining cost discipline across the organization. We’re very pleased with our progress so far this year.
We’ve diversified our business model in order for our earnings to be more predictable. And I believe you’ve seen that this quarter given the significant reduction in getting the loan sale revenues quarter-over-quarter.
Furthermore, we have the necessary liquidity, capital and asset generating capabilities to execute on our strategic plan and continue to deliver value for our shareholders. With that, I’ll hand it back to Sandro.
Sandro DiNello
Thank you, Lee. I am now going to close our prepared remarks with some guidance for Q4, and then open the call for questions and answers.
Please turn to Slide 17. We expect net interest income will increase slightly on an improved margin, partially offset by a seasonal decrease in interest-earning assets.
We expect gain on loan sale income to be flat. We expect return on the MSR to be between 6% and 8%.
And as we noted, we expect noninterest expense to be between $175 million and $180 million not including expenses associated with the Wells Fargo branch acquisition. This concludes our prepared remarks.
And we’ll now open the call to questions from our listeners. Jonathan?
Operator
Thank you. [Operator Instructions] We’ll take our first question from Kevin Barker from Piper Jaffray.
Kevin Barker
Good morning guys.
Sandro DiNello
Hi, Kevin.
Kevin Barker
Hi. I just wanted to follow up on some of the gain on sale margin pressure that you saw this quarter.
I believe there are some remarks, but I think it broke up a little bit. What was the primary driver besides some of the mix shift from delegating correspondent?
Sandro DiNello
There’s really nothing more to add to what was in the prepared remarks, Kevin. It was a combination of mix shift and some secondary compression.
So the mix shift is something that we felt was the right thing to do given the softness in the market and here is how I look at it. We could have moved $500 million maybe $1 billion of originations from bulk to retail.
If we’d have done that, you’d have seen GOS revenue and GOS margin be a little bit higher, but expenses would also have been higher. And importantly, our future risks to earnings would have been higher we haven’t taken down the high cost of retail or we could accept the lower margins, the lower GOS offset that impact with lower expenses or higher MSR return and higher loan servicing fees, still earn 13% on equity, 1% on assets, and having expense structure going into Q4 and Q1 that protects us on the downside.
So you can choose to focus on a mortgage revenue or you can focus on the stable earnings. We chose stable earnings, which I think is what you’ve been trying to get us to do since I first became CEO.
So margin is a black box, right. It moves around a lot depending on what the mix is inside of that black box.
And we don’t disclose the specifics of our GOS margin, never have, and we’re not going to start doing that today.
Kevin Barker
Okay, the expenses definitely did come down, which is a good sign, well below your guidance and you’re guiding to lower expenses going forward as well on a quarter basis. So, there’s obviously some protection of the overall earnings base.
Also, I just want to follow up on the attrition that you saw in the branches in regards to Wells Fargo transaction down $200 million. It seems like it’s in line with your expectations given that deposit base.
Has your expectation at all change going forward beyond the 17%? Or is there anything – when we think about the overall borrowing base – the customer base, do you expect any further attrition in the near term as we go into year-end?
And when we look at it in January, where do you think that number will sit?
Sandro DiNello
That’s really – you need a crystal ball to answer that question, but here are some general thoughts, and I’ll do my best to try to answer your question. First of all, as Lee said in his remarks, I think we haven’t had any attrition in the DCB transaction.
So that gives us a fair amount of confidence going forward that we are given the right things in this transaction. And as you know, we disclosed 17% post-closing attrition for this transaction.
Your company put out a paper on attrition last month that I thought was very good, very interesting. And it talked about the most successful branch acquisitions were ones where customer communication was strong and the branches had good proximity.
Our communication has been terrific, and these branches have terrific proximity to our branches. So I think looking at the composition of this deal and how we’re going about it using the same tactics that we used in the DCB acquisition gives me a lot of confidence that we’re doing the right thing.
Now this is a real different situation because, again as Lee noted, there are people that no longer live where they open their accounts. So what that’s can be going forward is difficult to estimate.
But going back to your paper, that paper talked about the average attrition happening between announcement and closing to be 18% in all the deals that were analyzed in that – in the analysis that you guys did. And that including close – closing attrition the total went to 29%.
So if you look at all those numbers and you compare them, so what we’ve experienced thus far and what we’ve been projecting going forward, I think that what we’ve said is reasonable and we’re still very comfortable with these projections.
Kevin Barker
Okay. That’s very helpful.
Thanks for all the color. I get back in the queue.
Sandro DiNello
Thanks, Kevin.
Operator
Thank you. [Operator Instructions] We’ll take our next question from Bose George from KBW.
Sandro DiNello
Good morning, Bose.
Bose George
Yes, good morning. Just first in terms of your guidance on the gain on sale margin increasing a little bit next quarter, is that mix or secondary market?
Or just any color on that would be great?
Sandro DiNello
That’s hard to be real specific about that, Bose. I mean, I think what we are guiding to overall in revenue was flat.
So how exactly that will come together that kind of detail I don’t think would be helpful, I will try to answer here. So our expectation at this point is that we won’t see much overall change in where the business is in Q4 and we are comfortable with that projection.
Bose George
Okay. Yes, that’s fair enough.
And then, actually, just switching to the spread. The net interest margin benefit from Wells Fargo, the 15 basis points, just given the timing of the close, is there sort of incremental benefit that comes in the first quarter as well?
Sandro DiNello
Absolutely. We’re only going to have one month of that this quarter.
We will have, obviously, three months first quarter, so absolutely.
Bose George
The – and, I guess, you guys had said earlier, I think, it was 21, is that kind of what we should think about overall?
Sandro DiNello
Yes. That’s still – let’s say, 20-ish.
So I think that’s right. And since we gave that guidance, we were still anticipating future – a December Fed hike.
Bose George
Okay, great. Thanks.
And then just one more on the Wells Fargo. Can you just talk about the expenses?
You mentioned the expenses in the fourth quarter going out to $10 million just from the transaction, but are there other expenses we should just think about from the Wells branch acquisition?
Lee Smith
Yes. And as I mentioned in my prepared remarks, you got the $10 million of one-time integration costs and then you’ve got one month of run rate because we’re going to close this at the beginning of December.
And so if you just look at my guidance, $185 million to $190 million, subtract the $10 million off, look at this quarter, add a little bit growth in subservicing, you can kind of back into what that one month run rate is. We’ve also previously published guidance on where the NIE and noninterest income of those branches will be.
Bose George
Okay. Okay, that’s good.
Thanks a lot.
Sandro DiNello
Thank you, Bose.
Operator
Thank you. We’ll take our next question from Chris Gamaitoni from Compass Point.
Chris Gamaitoni
Good morning.
Sandro DiNello
Hi, Chris.
Chris Gamaitoni
Hey, I want to start with the GreenSky announcement. Can you give us any sense of the characteristics from maybe an interest rate perspective on those loans and potential thoughts on the funding side?
How much more to come into funding?
Lee Smith
Yes. Yes.
Hi Chris, it’s Lee. On GreenSky, look, while we don’t expect volumes to be materially nature, it’s just another example of how we have diversified our business model and it can originate loans multiple ways.
And as I mentioned, it’s an efficient way given the origination in servicing of these loans is done via the GreenSky program, and these are when secured home improvement loans.
Chris Gamaitoni
Right, right, Lee, do you know the interest rates roughly that is maybe coming on it?
Lee Smith
No. I am just aware of the spread that we receive.
Chris Gamaitoni
And Lee, any other thoughts on kind of the outlook for loan growth at the community bank would be helpful?
Lee Smith
Well, we think we can continue the loan growth in the community bank, Chris, I will let Drew expand on that a little bit. Frankly, the results you saw in the Q3 are very comparable to the results that we’ve seen quarter-over-quarter.
Obviously, there’s different parts of the business that are stronger or weaker from quarter-to-quarter, but that’s the benefit of having a very diversified origination class month. So we can pick up in one area when it’s not so strong in another area, but let me let Drew add some color to that.
Drew Ottaway
Yes, thanks, Lee, and I think your point is well taken, and I think if you look at the third quarter, what we take a lot of confidence is the diversified mix of that growth. And so just like Lee alluded to, as we continue to look to different ways to grow the balance sheet, you can see our ability to do it in a really balanced disciplined way.
So while we’ve seen some of the quarterly growth moderate into mid single digit, we think we can continue to go out in that pace going forward.
Sandro DiNello
And I’ll just add. As well as the commercial loan growth we said on the – in the prepared remarks, it was a balance quarter-over-quarter and we think we can continue that the wealth of consumer loan growth and that will come from HELOCs, especially in mortgages we originate, we put on balance sheets as well as the indirect lending that we do.
So it’s coming from both the commercial bank and the consumer lending channels.
Chris Gamaitoni
Thank you so much.
Sandro DiNello
Thank you, Chris.
Operator
Thank you. We’ll take our next question from Henry Coffey from Wedbush.
Henry Coffey
Yes. Good morning, everyone, and thanks for taking my question.
Sandro DiNello
Hi Henry.
Henry Coffey
First, on some of your comments on capital. What is the move from CET1 being a restraint to total capital ratio being a restraint?
How does that kind of effect the things that are relevant to us?
Sandro DiNello
So it’s not CET1, Henry, it’s more a Tier 1 leverage. So as you can imagine under a Tier 1 leverage constraint because of non betas average assets, there is no differentiation between higher risk-weighted assets and lower risk-weighted assets.
Now that we’re risk-based capital constrained in total right now versus the constraint we do differentiate between the risk weighting of assets and there is a greater focus on the return on equity that we are able to generate from those assets. Then further I pointed in my remarks when I said about the focus of commercial deposits, the 31 basis point increase in our risk-based capital ratio from going out, pushing harder on those commercial deposits, brought a very nice capital benefit.
And while that’s not measured in the net interest margin, is measured in kind of the return on equity kind of the perspective.
Henry Coffey
So from a risk-based capital point of view, there are more things. You – our risk-based capital is more aligned with the way you would normally think about the bank in terms of making profitability decisions as opposed to all assets or all assets are bad or all assets go into the calculation?
Sandro DiNello
Yes. That’s a good way to think about it, Henry.
Henry Coffey
Assets are good. I don’t want to hurt anybody’s feelings.
Sandro DiNello
Yes.
Henry Coffey
So from the perspective of – you set a slip in there, you got a tremendous amount going on this year. We get through with some digestion for next year.
And this is not a new idea that you’re generating a lot of capital and life looks good and you’re building a great business. So what went – what trigger is the dividend of the buyback decision?
What sort of hard piece of wood needs the kind of hit the table to get that done?
Sandro DiNello
No. There’s a lot of things that go into that consideration.
First of all, we have invested in that, it’s about how do we get the best risk-adjusted return to our shareholders holding the long term, not in the short term. As long as we can keep doing what we’ve been doing relative to growing the balance sheet in a way that provides good return on assets as well as good return on equity, I think that’s what we should be doing, now there is other consideration for dividend, such as expanding the universe of shareholders.
So despite our ability to grow, it may make sense to do a dividend just because we can open up ourselves to more buyers. And there’s different consideration with share buyback, particularly when you still have a major shareholder.
So all of those factors and things where we think about on a regular basis. But at the end of the day, it gets back to, how do we get the best long-term risk adjusted return to our shareholders.
Henry Coffey
The other subject is getting into the minutia of the loan portfolio. So you have three areas of – that are relatively new.
The builder lending, which is seeing a lot of growth. Obviously, we’re also – the expansion in warehouse lending is a much simpler product, but the builder lending, the RV lending and now the GreenSky program, how do those – what are the individual – what are the expected kind of loan yield and loss dynamics on those loans?
And how do – what’s the set of franchise value of a home equity – of a home improvement loan or an RV loan? How does that fit into kind of your customer relation strategies?
Sandro DiNello
I can’t speak to franchise value, I guess that’s for the market to determine how it impacts franchise value. What I think it does by having these various sources of loan production and what I think is unique about it is, as a midsized bank, we’re not limited to what we can get within the footprint that we have our bank branch offices.
So we can go outside of that footprint. We make sure that we hire the right people that have the right relationships and can get the right kind of assets on our balance sheet, and we can continue to grow the net interest margin and the net interest income the way we have at a 15%, 20% clip for a number of years.
If we keep doing that and we’ve done it in a way that doesn’t exposes any large concentration in any one area to me, that’s the build franchise value as opposed to what the franchise value in RV loans versus the franchise value of the builder loan or the risk of one versus the other. I firmly believe that if you’re making loans to the right people, it doesn’t matter what kind of loan it is, you have similar credit performance over longer periods of time.
And I am going to stay away from concentrations where it probably one geography or one business could cause a significant impact on your results, I think that’s a business model that works.
Henry Coffey
Great, thank you.
Operator
Thank you. We’ll take our next question from Scott Siefers from Sandler O’Neill and Partners.
Scott Siefers
Thanks for taking my question. I want to try to take a second stab at the gain on sale margins just given there was a surprise this quarter.
I mean, what is that leads you to believe that it will indeed increase as we look into the fourth quarter? I mean, are you planning to shift around the mix which you guys could do consciously?
Or are you seeing any relief in the competitive dynamic?
Sandro DiNello
So let me first address your question about shifting around the mix. So we’re not – we don’t view this as what the margin is.
So we look at this as what’s the revenue. What’s the net revenue to the company.
And so we’re going to adjust where the business comes in based on the overall net benefit to the company. And what we’ve been trying to do is build and I’ve been saying this for a long time, build the bottom, right, what’s the bottom that we could experience and make sure that, that isn’t so low that causes us not to get the kinds of returns that we need to.
And I’m talking about 12% to 13% on ROE, 1% on ROA. So this is a dynamic thing.
We manage it from week-to-week. I don’t know where the business is going to come from next week or in the 30 days.
If you think we’re that good, we’re not. But what we’re at good at is adjusting to where the opportunity is.
And so at the end of the day what we are telling you is, we’re confident that the overall gain on sale income can be flat over the course of this next quarter. And given the expense structure that we have in place now and the expenses that we guided you to, we believe that the – we can continue to have the return – overall returns on equity and return on assets that we are committed to.
And that’s only happening because of the growth in our banking revenue and the growth in our servicing business and our ability to offset some of the challenges in the mortgage business with enhancements amidst our returns as well. So it’s just different.
I can’t be that specific with you. It’s not that simple.
Scott Siefers
All right. Thank you.
I guess, next question is just on the cost side. So is there any – and, I guess, this is part of why the gain on sale becomes important because, presumably, there is profitability to – like it’s one thing if costs are going up because you’re originating more retail, for example.
But I guess, I wonder if a relatively greater proportion is, say, correspondent, is there a reason to believe that you could have additional flex on the cost base as we look into 2019 as – if should things not get any easier in the gain on sale environment?
Sandro DiNello
Well, sure. We have always adjusted our cost base to what we need to based on what the revenue opportunity is.
So we’re – there is never a bottom side to what the costs are. If you say the business doesn’t work and you just shut it down, right?
So we’re not – we will not find ourselves in a position where we’re not able to adjust. There’s sometimes – I can’t tell you that it’s never going to be a timing issue going forward.
I can’t predict the future. But I think our history shows that we’ve done a pretty damn good job of not letting that get in the way, right?
Look at this year, we started adjusting the cost of our business way back in January, maybe we didn’t notice, because we didn’t know, but we’re preparing for just exactly what happened. And so despite the fact that we had a decline in revenues this quarter we were able to have a similar decline in expenses, offset by additional revenues in servicing and MSR, and that doesn’t just happen.
Those are things that we plan for. And we’ve been right on it, I can’t guarantee we’ll always be right, but I think our track record did present that.
Scott Siefers
Okay, thanks. And I guess, final question is just on MSRs and selling a descent chunk.
So I guess, it looks like probably at a higher watermark for the net return on MSR assets. Can you just walk through your thinking on just basically what’s going on there?
Lee Smith
Yes. So when you look, Scott, its Lee, if you look at the return on the MSR asset, it’s a combination of factors that has led to the healthy return.
And we’ve seen low runoff in another quarter where there was strong hedge performance, but it was a high carry, because we’re holding a large remisier asset. The CPR was the lowest it’s ever been.
It’s 5.6 for the quarter, I can’t remember it being lower than that. And a lot of the MSRs, pretty much all the MSRs that we have in our portfolio are newly minted, so there is less chance of them paying off, and so that leads to strong evaluations.
I mean, in Q3, we did see some positive transaction economics in the quarter, which included collection of some of those call backs. So it’s a combination of factors.
These aren’t necessarily one particular thing.
Sandro DiNello
Yes, I like to add here that what we built here over time is something that’s pretty extraordinary. We now have relationships with funds and investors that buy MSRs.
They’re now preferred to buy these bulk MSR with others. The performance of the MSR has been solid.
The way our servicing team has been able to provide these recapture services and all the other things that go around quality in the servicing business is really important. And so this growth in the number of loans that you’ve seen us service, I mean, we’re going to be at 800,000 by the end of the year growth this year.
It doesn’t get a lot of attention, but it’s huge. And it doesn’t come just from the Flagstar loan, the generation we are now, scheduling volumes on a quarter-by-quarter basis, I think we’re booked up for quite a few months in terms of the number of loans that we can onboard from outside sources.
So these MSR partners that we built relationships with are so satisfied with the performance of the MSRs that they buy from us that they’re now bringing their servicing business from other services, so this is growing much faster than we thought it would. And this – the fee income that’s growing from this business helps when we’re in difficult times here for the mortgage origination revenue.
So these are all things that are coming together and it’s the plan that we’ve been focused on building for a long time.
Scott Siefers
All right, perfect. Thank you.
Operator
We’ll take our next question from Daniel Tamayo from Raymond James.
Daniel Tamayo
Good morning, guys. So I guess, building off that MSR discussion, I think Lee mentioned that you had about 200,000 accounts come on in the third quarter.
And that led to strong custodial deposit growth. I think you’ve had about $400 million there.
What is the beta on those deposits? And is there still a reimbursement of spread required on those deposits?
I’m thinking back to the company control deposits discussion we had a few years ago.
Sandro DiNello
So two things, Danny. One is, when you’re dealing with those deposits, there’s no us, and you’re absolutely right.
There is some arrangement we have with the MSR owner to pay down, but it’s also that whatever benefit we pass back to the MSR owner is part of the whole mix with how they pay us for servicing, as opposed to beta, because they have to pay interest per se for that deposit, but it is a cost that has a beta that is tied to, it’s baked in, yes, it’s right, it’s baked into all the service agreements.
Daniel Tamayo
Okay. And would you say that you’re expecting similar rates of growth, I mean, with 200,000 accounts?
I think I got the quarter wrong. The 200,000 accounts expected to come on in the fourth quarter, are we – should we be looking at another acceleration in custodial deposit growth in the fourth quarter?
Lee Smith
Yes. So it’s Lee.
Yes, the 200,000, I was going to chime in at the end, it is the fourth quarter. And these are 200,000 non-Flagstar originated loans.
So that’s when the growth is coming on. And obviously, with higher loan count, you can expect to see high custodial balances.
Sandro DiNello
Over the long term, there will also be seasonal headwinds with those balances this quarter. But generally speaking you’re thinking about this the right way, because the number of loans we service grows, the sort of deposits grow in a similar fashion, but the one factor that, that changes over time is other than the seasonality within the year is refinance activity.
If there’s a lot of refinance activity, which would bring a lot of payoffs than your custodial balances will increase in that usual way, but we’re not seeing growth in refinance activity. So I don’t think that becomes a factor as you try to calculate what the deposit – what the custodial deposit growth might be.
Daniel Tamayo
Okay, thank you. And then, I guess, my last question on that topic just in terms of now you’re at – you’ll be over 800,000, what kind of growth can we expect to continue to see here going out into 2019?
And maybe, I mean, next year, what are you thinking – what kind of infrastructure do you have in place to handle over 800,000 loans?
Lee Smith
Yes. So we only guide a quarter ahead.
We don’t give any further guidance than that. And as you can imagine, often we are onboarding loans.
These – a lot of these are bulk deals. So they are not – they’re not necessarily easy to predict.
But what I’ll say and I’ll come back to the prepared remarks, we’re very close to having a capacity of two million loans. And so at 800,000 by the end of the year, we wouldn’t be 50% of what the capacity is when we were at 2 million loans.
And so at 800,000 by the end of the year, we wouldn’t be at 50% of what the capacity is when we were at 2 million loans.
Sandro DiNello
Let me add. The capacity will not be an issue.
We will have the capacity ready when we need the capacity. So don’t let that capacity issue factor into what if I think our loan totals might be.
Look at our growth in the past and figure out what you think it might be in the future.
Daniel Tamayo
All right. Terrific that’s all I have.
Thanks guys.
Operator
Thank you. We’ll take our next question from Scott Beury with Boenning and Scattergood.
Scott Beury
Good morning, guys. Just most of my questions have been answered.
But I just wanted to ask something about the loan yields and the alternative investment portion. You saw a pickup of about 12 bps linked quarter.
Just curious if you had any thoughts or any color on what the trajectory is for repricing in that part of the portfolio? Obviously, it’s going to be dependent on mix and things of that nature, which I’m sure you can’t totally project, but generally, does this seem like the delta here linked quarter?
Does that seem like a decent run rate all things considered looking at what we’re seeing for rate hikes in 2019?
Sandro DiNello
I think – Scott, I think the delta could even be somewhat muted. There were some payoffs in this quarter and payoffs in last quarter.
So that effects the comparability between the two quarters. But as we look at the commercial portfolio, high percentage of that is variable.
So as we continue – if we have the rate hike at the end of the quarter, I would expect a full benefit and that I am shining through next quarter and would also point to continue to manage our deposits to a very low beta. So I think that we could expect additional expansion next quarter just from that phenomenon.
And it all depends on where the loan mix is, but also where the old loans pay off at and where new loans come on at.
Scott Beury
Okay. I understand that.
That’s very helpful. Yes, most of my other questions have already been answered.
So thanks for the color.
Sandro DiNello
Thank you.
Operator
Thank you. Ladies and gentlemen, there are no further questions in the queue at this time.
I’d like to turn the call over to Sandro DiNello, President and Chief Executive Officer. Please go ahead sir.
Sandro DiNello
Thanks, Jonathan, and thanks to everyone, for your interest in Flagstar. The common theme all three quarters this year has been the softness in the mortgage market, not just for Flagstar, but for all originators, yet our story every quarter has been that despite the mortgage challenge, we have posted strong consistent earnings.
It is now clear that our banking business is carrying debt. We have grown that business to the point where we don’t need to rely in mortgage business to produce strong returns on assets and equity.
Looking ahead, we believe we’re well positioned for continued success as we grow our banking business further. Our pending acquisition of the Wells Fargo branches will significantly increase our core deposits.
Further, it gives us presence in markets that are continuous to our Michigan branch network, opening opportunities for other banking services, including commercial and consumer loans. Additionally, as I mentioned in my opening remarks, the lifting of our Federal Reserve Supervisory Agreement ushers in a new era by providing more freedom in capital management.
While we continue to believe we have attractive growth opportunities, we expect to consider other options for excess capital, including share repurchases and dividends. Overall, we will pursue a strategy to maximize these, as I said, risk adjusted returns for our shareholders.
Lastly, we appreciate the loyalty of our shareholders in their support of our business plans, and we thank our employees for their tremendous work and the pending Wells Fargo branch acquisition, and all they do every day to make Flagstar a success. Thank you for your attention this morning.
I look forward to reporting Q4 results in January.
Operator
Thank you. Ladies and gentlemen, this concludes today’s teleconference.
You may now disconnect.