Mark Seaton - Executive Vice President and Chief Financial Officer
Craig Barberio - Director of Investor Relations
First American Financial Corporation (FAF) Barclays Capital Global Financial Services Conference September 10, 2013 3:30 PM ET
Good afternoon. Thank you for joining us. Very pleased to have team from FAF. With us up here on the podium is, for those of you don't know, I'm sure most of you do, FAF is second largest title insurer in the country.
Speaking today will be Mark Seaton. Mark is the CFO, was appointed to that position this year. Mark's been at FAF since 2006, prior to that he spent some time in private equity and in investment banking. He's also joined up here by Craig Barberio, Head of Investor Relations.
So with that quick intro, I'll hand it over to Mark for his comments.
Thank you, Mark. I want to thank Mark and Barclays for hosting us here. This has been -- I don't how many years in row we've been at this conference, but there's always a good turn out of lot a good companies, and we've had a very busy day of meetings today, so I want to thank you for inviting us here.
I'll just give a quick overview on First America. We've been around for 124 years, that's a lot of real-estate cycles. We started back in 1889 in Santa Ana, California. We still have our headquarters there to this day. We've got a significant presence in U.S. We're also the largest provider of title insurance internationally. We've got a business in Canada, the U.K. and Australia and some other countries that we'll talk about later. And we do residential and commercial title insurance, that's our core business.
We've got a strong competitive position. 26% market share in the U.S. We've got one national brand that we distribute our products through, and we distribute our product both direct and agent. So on the direct side, we've got the relationship with a customer that might be a real-estate agent or broker or lender. We do the title search and examination and we issue the policy, and we keep 100% of premium.
On the agency side, independent agents of First America or attorneys or separate title companies, they have the relationship with the customers, they do the search and exam, but we issue the policy, and in return we keep about 20% of premium for that. So we've got extensive distribution network, both on the direct and the agency side.
We've got industry-leading technology infrastructure and title plan information, which I'll get into a little bit. And from a financial perspective, we do about $4.5 billion of revenue, very conservative balance sheet, very strong conservative investment portfolio and low financial leverage.
So in terms of our organizational structure, our core business is title insurance. 93% of our revenue comes from title segment. Only 7% comes from the specialty insurance segment. On the title side again, it's residential, commercial, international. Database solutions is really data that we have to help us with our underwriting decisions. And the real-estate services and trust and banking are just complementary businesses that we have that are synergistic with our core title business.
And then, on specialty insurance side, we've got really two businesses, home warranty and property and casualty insurance. Home warranty is used in most states, and typically the seller of a residential real-estate property will typically buy a home warranty contract for a buyer of a house to serve a good faith gesture, to make sure that the air conditioning system is working and the refrigerators working, it's simply a one year contract.
We're the second largest home warranty company, behind a company called America Home Shield. We get a lot of claims in the summer, when air conditioning units go out and we don't get a lot of claims in the winter time. So it's very weather dependent, but overall the combined ratios in home warranty are mid-to-high 80s typically.
On the P&C side, we don't take a lot of underwriting risk. We don't insure within 1,500 feet of a breadth zone. We're primarily homeowners insurance. We do a little bit of auto, but the vast majority is homeowners insurance in the southwest, California, Arizona, Nevada, et cetera, and the combined ratios in P&C are also about mid-to-high 80, so very strong.
So going back to the title side. Our residential title business, we have 800 offices and most of the county, and think about it kind of like the county branch network, if you're going to go your local title company in your local county, that's where our offices are. So we've got 800 offices. We've got a significant agent channel. We've got 7,500 agents in the company. Most of those agents are on the East Coast. We're more direct as the further you get west.
And we distribute our product on the residential side in two basic ways, centrally and distributed. The distributed are 800 offices, but we also distribute title policies in a centralized manner. So really our centralized business sells for the top 20 lenders directly. It sells to services and trustees and a default client directly. That's really our residential business. That's really the core, what we do.
On the upper right, our commercial business, we're in about 40 commercial markets in the U.S.; L.A., New York, Chicago, San Francisco, Dallas, et cetera. All 40 of our offices are meaningfully profitable. We're seeing a lot of commercial business now, lot of office properties, multi-family, both purchase and refi. So we're really seeing a lot of momentum in the commercial business.
And the fourth quarter of last year was the best quarter we've ever had in commercial. And so far for the first six months of this year in 2013, our commercial revenue is up 27% over last year. And the commercial market in order to be stronger, you've got to have good ratings, you have to have a good balance sheet.
A lot of our customers really care about counterparty credit risk. So they really examine the counterparties. And fortunately we have a good story to tell there. And we see continued growth in the commercial market for the next few years. The margins on the commercial side are much higher than on the residential side.
International, we're primarily focused on Canada and the U.K. We've got a meaningful business in Canada. International is about 10% of our revenue and most of that is Canada. There is five or six banks that do most of the mortgage lending in Canada, we've got great relationships with all of them. A lot of those banks were here presenting at this conference today. So we have a meaningful business in Canada, a smaller extent to the U.K. and also Australia. So we're really focused on the British Commonwealth countries that have established legal systems and are driving mortgage market and it's all direct, there's not a lot of agency business is done internationally.
And finally, in database solutions on the bottom right, it's really our title plant business. So there is 500 counties that we own a title plant in. And whenever we get an open order on the titles, we go into our title plants to get the appropriate title information. We look at who owns the property, what the legal description is, what the leans are that are attached to that property, whether it's in a incumbent sort of a mortgage, and we use that information to make our underwriting decision.
And so our title plants, we're in 500 counties that represent about 70% of the transaction that are done in the U.S. Our biggest customer is our title company, our First American inter-company, but we also have all the major underwriters as customers and a lot of agents as customers, and we sell this as kind of utility for the industry. And we have a largest database of this kind in the U.S.
We're a leading market position. We've got 26% market share in the U.S. Fidelity is number one, they've got 34% market share. We used to be the largest, up until about 2008, when Fidelity bought Land America and they sort of vaulted into the number one position. But after that Old Republic is the third at 14%, Stewart at 13% and it really falls off after that.
We've been in 26% market share for probably five or six years now, which just hasn't changed, it's been 26%. But it's been that way, despite the fact that during that timeframe, we've already closed half of our offices. Six years ago, we had 1,600 offices, now we have 800 offices.
When you get out of an office, you're intentionally relinquishing market share and you're doing that because that's the profitable thing to do. We've also terminated about 3,000 of our agents that just didn't fit into our network. So our market share has been flat, despite the fact that our office footprint has been cut in half and we got rid of 3,000 agents.
And so right now, a lot of the efforts that's being done is to gain organic share. And we're trying to gain that in the top 50 counties, many of which are in the top states, as you can see listed here. So here's our market share broken out in the top 10 states. And we're really under-represented in the California, Texas, Florida and New York. And that's where a lot of our market share focuses, and we're trying to focus on recruiting responsibly and using technology to help us gain market share and that's where a lot of effort is being made right now.
In terms of just the current market environment and there's a lot of talk here at the conference about what's going on with the purchase market and the refi market, and we'll just kind of give you our sense. In terms of the purchase market, I mean we feel like we're on the early stages of a long-term recovery in the purchase market. We've got sales activity and housing prices are improving.
For the first six months of this year, our closed transactions on the purchase side, forgetting about refi, just on the purchase side, are up 18% versus the first six months of last year. The transactions are up 18% and the average fee that we get for every purchase transaction is up 10%, mostly because of rising housing prices, the fee that we charge as a function of the housing price. Whenever housing prices rise, we get a natural fee increase, but also because of a greater mix of higher and more profitable areas.
So our revenue on the purchase side for the first six months is up 28% over last year. And we feel like they're going to need several more years of growth like that to get back to the normalized trend line. And for us, one of the things that very important, when analyzing our company is we get paid twice as much for a purchase transaction than a refi transaction.
Our average fee for a purchase transaction is about $1,750. Our average fee for a refi is $800. So we like refis and we'll take them all day long, but we care twice as much about the purchase market. And that business, although depressed is growing and we feel like it has a lot of legs to it, a lot of tailwinds.
We had a strong spring selling season that was one of the big question marks going into 2013, as how does the spring season look. Typically, and I'll show this on the next slide, but typically May is the peak month for opened purchase orders. May is the peak month. Some times it's April, some times it's June, but most years it's May.
Well, this year in 2013, our purchase orders in May, June, July and August were all the same, about 1800. So we had this peak that just sort of lasted much longer than it typically does. And now that we're heading into the fourth quarter, obviously we'll see just a normal seasonal decline in the purchase for just as people buy fewer homes in the winter time. But the spring selling season was strong and it actually carried into summer, which is good. So we had a strong spring selling season.
Interest rates are up. The 30 year mortgage rate is 4.75%, the last time I checked, but it hasn't affected the purchase market, it's affected the refi market considerably, but not really the purchase market. We've got inventory shortages in many markets. That's one of the reasons why you're seeing price increases. And foreclosure activity is moderating, especially in the western states, where there are really trustee-driven states. A lot of foreclosures are back to normalized levels. It's really on the eastern states, where the court sounds foreclosures, where you're still in elevated foreclosure activity.
On the refi side, it's a different side. In refi orders are really drying up, if not, have dried up. And April was really the peak in terms of our refi or as we opened about 3,300 orders a day in April on the refi side. In August, we opened about 1,600 a day. So our orders are down about 50% on the refi side since April and I don't think it was a surprise that they were going to fall. I mean everybody kind of knew that refis eventually were going to start drying up. I think obviously, that's happened and really what we're doing now is backing aggressively on the cost side to reduce our expenses accordingly, given that refi drop.
We also have continued strength in the commercial business. I kind of referenced commercial revenue being up 27% year-over-year for the first six months of this year and we see continued strength in the commercial business. There is a lot of capital on the side lines and when we talk to our commercial customers, we feel like there is meaningful growth still left in commercial. And again, we have a disproportionate share of the commercial market share.
This Slide shows our open order account trends. The top gray line is our total orders per day going back for the last 13 months. You can see there's been a general decline, really because of the refis. The dark blue line in the middle there is our refi open orders per day. The red line is the 10-year treasury rate.
So one thing I'll point out, as you can see there is really a very tight correlation between what happens with the 10 year treasury and what happens with our refi orders. And you can see in April, as the 10 year started to increase, we saw almost a one-for-one decline in our refi orders. So refis are very, very sensitive, as you know to rates and this slide shows that.
The bottom line is our purchase orders per day, and it's kind of hard to tell here, but again we're seeing 25% growth year-over-year in the purchase market roughly. And as of July and August we're opening more purchase orders than refi order. So you can see the number, the lines there start to inflect the last month or two, which is something we're starting to anticipate a while back. Normally, about 30% of all transactions are refi-ed. It's been running elevated for the last couple of years.
In terms of our margins, this is a graph of our title insurance margins. We're adjusting for two things. We're adjusting for realized gains and losses in the investment portfolio and we're adjusting for reserves for anything for which we had an $89 million strengthening last quarter. But this sort of gives a best picture in terms of our operating earnings for the company. The blue line is our pre-tax margin, which correspondence to the right access and the grey bars are the market from mortgage originations, which correspondence to the left axis.
And you can see that we've only been in 10% margin lines, really three times going back to 2000. We did it in 2003, we did it in 2005 and we did it in 2012, last year. And in 2003 and 2005, we were in a market that was really $3 trillion to $3.5 trillion. And we also did it last year in 2012, in a market that was about $1.75 trillion.
So here you had a market that was really half of what it was during the boom years, and yet we had close to record margins. And the reason why we were able to do that is we just took a lot of structural cost out of the business. We worked hard over the last few years to consolidate our accounting centers, our claim centers. We've gone from a 100 claim center to four. We've gone from 30 accounting centers to three. We've gone from 30 datacenters to two. We've collapsed a lot of other structural costs out of the business that have allowed us to reap those margins.
And when we look forward and we get to a more normalized market with a mix of closer to 70% purchase, we feel like there is room for improvement in the margins from these levels. So far in June for the first six months in '13, we're at about 11% margin, similar to where we were in 2012.
This is just a snapshot of our quarterly financial results. You can see on the upper left our revenues by quarter. The grey bars are 2012, the blue bars are 2013. You can see for the first two quarters at least we've got revenue growth over last year, driven by both strength in the refi market and strength in the purchase market and commercial. Really all three are higher than they were last year.
Pre-tax income on the upper right, you can see the second quarter of '13 was lower, really because we took that $89 million reserve strengthening. If we didn't have that, our pre-tax earnings would have been about $148 million and you would have seen that in our earnings growth.
Just a quick comment on the reserve strengthening. We had adverse development for title claims, really for policies '04 to 2008, so our title claims are coming down. In the second quarter, they just want to come in down at the level that we anticipating, and therefore we have to take a true-up. But as we sit here today, we're at a very conservative estimate within the range of reasonable estimates and we feel very confident with what our reserves are.
You kind of never know what reserve, the area that has a biggest judgment in our accounting, but we feel confident that we're appropriately reserved. We've told the Street on our on our last earnings call that we're going to book at a 5.8% loss rate for the second half of this year. And I think there is a greater likelihood that that's going to trend down over time rather than up. And you can see our cash flow, which is something that we really track, is on the bottom right. And our cash flow in the second quarter was significantly above, where it was in the second quarter of last year.
Our balance sheet is strong -- strong balance sheet. On the asset side of the balance sheet, it's very liquid. We've got $860 million of cash. That's not all at the holding company. We have about $175 million of cash at the holding company, that's I would say free cash. The rest of it is in different subsidiaries that we have. We have a bank called First American Trust that holds a lot of cash for liquidity purposes. But about a $175 million of that roughly is at the holding company, that we'd consider free.
We've got about a $3.1 billion investment portfolio. On the liability side, we've got demand deposits of $1.6 billion. We manage anywhere between $4 billion and $5 billion of extra deposits, third-party funds. When somebody puts a 3% down payment on the house, well often times, First American will manage that money.
Most of those funds we get to third-party banks. Some of it we give to our own bank, First American Trust. So we did give it to our bank, then it goes on balance sheet and that's what those demand deposits are. They are really escrow funds that are on our bank and therefore on balance sheet. We also have reserves of $1 billion, which is almost all title insurance reserves, debt of $315 million and equity of $2.3 billion.
So ROE of the last 12 months has been almost 12%. We've got 11.8% debt-to-capital ratio as of June 30. The normalized debt-to-cap ratio for First American has been around 20%, somewhere in the low 20s. We've been considerably lower than that, at least since the spin-off and we don't have a perpetual philosophy to run the company at an ultraconservative debt-to-cap. I think at some point we will bring it up to the 20% mark and we're just waiting for the right opportunity to do that. And we've got statutory surplus of $1 billion. This is in a primary insurance company, which is called First American Title Insurance Company.
So just to conclude here investment considerations, we're in pure play in the title and in mortgage markets. We're not in exotic businesses. We've got all of our eggs in the title insurance and in the mortgage space. So we feel like we're a very much of pure play. But also a pure play that's leveraged towards the purchase market, and just to reiterate, when you pay twice as much for purchase transaction and we feel like we've got short terms headwinds and the sense of refis are really drawing out, but long term, certainly in 2014 and beyond we're going to have a lot of tailwind in terms of the purchase markets, so we're very leveraged in that arena.
We've got a strong competitive position and we're pursuing growth opportunities, both internationally and domestically. Most of our growth strategy revolves around organic opportunities. We will do acquisitions. I think we'll probably do more acquisitions over the next five years than we've done over the last five years. But we're very selective in terms of the opportunities there are and to do a transaction at First American is very high right now.
We've got structural cost reductions that have really enhanced our earnings power. We've taken out a lot of cost that just aren't going to come back when the market normalizes. There is always cost that we're going to need to add when volumes rise, but there is a lot of fixed cost that we just don't ever see coming back. So we feel like the earnings power is higher now that it has been in quite some time, particularly as you think about the loss ratio eventually turning down.
That being said, we're always looking for ways to cut structural costs. This is always something that we're focused on. We got a very good financial position and we have a meaningful opportunity to return capital back to shareholders. Last year, we entered the year, paying a dividend of $0.06 per share per quarter and we doubled it throughout the year, so now we paid $0.12 a quarter, $0.48 a year.
Our yield is 2.2%. We feel like there is opportunity to raise the dividend overtime as our earnings improve. And we also repurchased about 1.6 million shares last quarter for $36 million. We've got a $150 million share repurchase authorization. And we haven't really used it too much since we've been a public company for three years. We did the spin-off of Core Logic. But last quarter we felt like our stock pulled back to the extent that we felt like it was attractive. And so for the first time we sort of jumped into the share repurchase mode for a total of $36 million. So we do think there is a meaningful opportunity to return cash to the shareholders, either through dividends or buyback.
And I want to thank everybody for their time.
All right, that brings us to the audience response section of the presentation, so for those of you who want to participate, please pick up your handheld device. First question, please?
If you currently don't own shares of FAF for underwritten stock, what will cause you to change your mind: one, stability and rates; two, robust purchase volume; three, lower valuation; or four, increased capital returns?
So nearly 50% -- 47% indicated robust purchase volumes, with the next most popular, 29% at lower valuation.
Next question. In your opinion what is the best incremental use of capital for FAF: one, increase dividend; two, increase buyback authorization; or three, M&A?
So 44% indicated increased buyback and 39% increased dividend with only 17% looking for M&A.
Next question. Over the next six to 12 months what do you expect will happen to your exposure FAF: one, increase; two, maintain; three, decrease; or four, remain uninvolved?
All right, of the 56% involved, 39% all goes to increase and 11% maintain, only 6% decrease.
Last question. What causes the most risk to the investment pieces over the next yea: one, increase provision expense; two, decline a refi volumes; or three, lower title margins.
Okay. So it's evenly swayed between the decline in refi and lower title margins, both at 44%, only 11% concerned about the provision line.
So thank you for your responses. I'll kick off the Q&A and then we'll open it up.
Mark, you indicated towards the end of tail end that you expected to do more M&A than you had at over the last five years. Can you just talk a little bit more about what you would like to target and what's the ideal candidate is? What type of businesses you might say within that?
Yes. I mean the logical targets for us are just title agencies and title underwriters in our space, but frankly there is just not been many of them, at least not have done many immaterial ones, but we'd like to do more material transactions than small tiny little one 3% title agencies.
But there is always opportunity in the title in our core title space. More and more there is opportunities that are rising just in -- and I just call it mortgage processing or mortgage analytics, other products or services that are sold to health vendors, process mortgages. And so I think that's an area that we would look at, not insurance type companies that have risk involved, but more just labor intensive processing type companies. So those are some areas that we've been looking at lately.
And as you look to acquire potentially in the mortgage processing businesses, will there be any concern of not wading into the waters, you previously in with CoreLogic and not competing in that area or would you -- is it possible that you would add more businesses that would directly compete to that.
Well, I think CoreLogic did a lot of things. I mean we don't have the strategy of just competing and trying to be the next CoreLogic. I think in series there is, areas that we could overlap in, but that would just be on the margin. I don't see us going head-to-head. I mean there -- they've got phenomenal data, phenomenal [indiscernible], so we're not trying to completely replicate that. But I think there is ways that we would compete with them, but really on the margins, not head-to-head.
And how do we think about and quantify what your excess capital is that you have to return. I mean you guys don't have a notion of enforce and clear capital requirements, help us try and quantify what's available to return or either return or deploy over the next couple of years.
Well, we've got again $175 million of free cash at the holding company. And that's all just excess capital. Now, we'd like to have about one year worth of cash obligations at the holding company. A one times just to pay our dividend or interest expense, and we have holding company expenses. That's about $100 million a year. So right now we've got $175 million. So we've got $75 million more than our target really.
We also have $1 billion of surplus at our insurance company. Our goal at the insurance company is to maintain A-minus financial strength rating, that's the goal. It's not like we have a premium surplus goal or an amount of surplus. It's really the surplus that we feel like we need to maintain A-minus ratings with the different rating agencies. And different rating agencies care about different things. Some care about debt-to-cap, some care about risk-based capital, some care about premiums of surplus. We probably have a $100 million to $200 million of excess capital at the insurance company. That we don't really need and we can still maintain our A-minus rating.
And then the third bucket I would say is just our debt. I mean we're at 12% right now, put some debt-to-capital. We can go up to 20% comfortably. Our covenant says we can't go above 30%, but we can go low 20s very, very comfortably. So you can do the math and kind of figure out what our excess capital there is.
The other thing I would say is that when we did our spin-off three years ago, there were reasons why we wanted to be conservative with our balance sheet. We were new company. The purchase volumes were declining at that time. Our margins were 6%, not really where we wanted them to be. Our claims were still rising and we had other risk out there in the marketplace.
Now, we've been three years of the public company, purchase volumes are increasing. Our margins, although, they can always do better, they're significantly higher now than they were back then. And our claims were coming down. So now it's not like we have a need -- I mean we always want to run the company with some level of conservatism, but the risk that we had back then are all gone now.
And so I think you'll see us to be more aggressive. And that's one of the reasons why you start to see us buyback stock last quarter and why we doubled it back, because we're feeling a lot more comfortable about the environment now.
How does the plans -- the acquisition of LPS by Fidelity, how does that change economics of different aspects of your business?
It doesn't materially affect our business. I don't know if it changes the economics of the business, although I think we're optimistic that we'll pick up a little bit of a share that comes out. I mean, we said we went to title covers get together, they don't really keep all of that share. I know LPS did a lot of other things outside of just title insurance, but they did have 3% of the market share.
And so we're optimistic that we can get at least a little bit of that. But other than that I think it was a probably a good deal for Fidelity. It's going to be probably a good thing for First American too. But it's not going to be -- I don't think it's going to be material anyway, just on the margin, probably could change the economics of the title business at all.
You had mentioned that your market share stays flat over the number of years, but if I look at Slide 9, it looks like your purchase open orders have remain constant in the phase of steadily decline purchase applications, so are you gaining market share or what is the time out?
No, I mean I think there is disconnect between our purchase orders and maybe some of the applications that you're saying. I think if you looked at our purchase orders versus our competitors, you're going to see a similar trend. I think we're gaining a little bit of a share, but it's less than 50 basis points. I don't think its material enough to show up in this number. I think there is just a disconnect between what you see on like [indiscernible] versus our orders for some reason.
Just a couple of basic questions. When you quoted the fees per loan, is that just a premium for the policy or does it include like other cases where you do the settlement work and add some additional fee, is that kind of an all-in revenue number or is the revenue above that?
It's really all-in. It includes the title policy, the fee we get from the policy as well the fee we get for the escrow in the closing, so it's fully those two fees in one.
And so a basic question, on the commercial side, isn't it the same dynamic as in the residential side where it's driven by, mainly by lending against commercial real estate as opposed to just any old commercial real estate transaction or what are the current market dynamics there?
It's both. If you've got a principle who is a buying a property that's going to be title fee, whether it comes with a mortgage or not, so it's similar on the residential side in a sense that there is purchase transactions in the commercial and there is just refi transactions in the commercial. Maybe the difference is on the commercial side, there is a higher frequency of refis just because you don't have 30 year loan to commercial typically of five years or seven years. So you have a higher velocity of refis, which is obviously a good thing for us.
Can you talk a little bit about what kind of sensitivities do you see on the commercial side to the way it's moving up?
I hope I answer your question here. The sensitivity on the commercial side, I mean, going into the year, we were expecting sort of mid-single digits growth from commercial, and so far this year, we've gotten 27% growth. Now, again I don't think that rate is sustainable for the second half just because the second half of last year was so strong. The fourth quarter was the biggest quarter we've ever had.
So we do think that there is going to be incremental growth in commercial. And you know there were a lot of CMBS deal that were done five year ago, that needed to be refi-ed, so we have a refi pipeline. There is a lot of capital out there on the sidelines, just looking again commercial real estate and it's -- we're seeing it very broad based. All of our 40 markets are strong.
It sound like it's happening in one area. And we're seeing it across different assets classes, like in offices, multifamily, so it's -- the broad base strength gives us a lot of comfort. And we haven't seen a lot of sensitivity with the recent move in the interest rates, that hasn't really affected the commercial business, at least as of yet, like it has on the refis side. So that's all good news for us.
Can you just talk about the difference in manpower per se versus purchase for the refi. We need to hire more people if purchase picks up. And then second question would be, can you just update about title margin where you think it could go. Obviously you guys are heavily regulated, fees can come out and kind of change the fees that you guys produce, what do you think that upper limit would be before you go California against your something in that nature?
On your first question, it takes more work to do a purchase transaction than a refi because a purchase transaction, the property is actually changing hands and therefore there a lot more background checks that we have to do as opposed to refi where it's not changing hands.
And so we kind of talk about the gross margins of purchase and refi being about the same. So meaning that the personnel expenses, the operating expenses for the purchase transaction is twice is much as of refi transaction. But the profit dollar is twice as much and therefore we can spread that over our admin cost a lot better. So we must going to have a strong purchase market than a refi market.
The second question in terms of where the margins could go, it's interesting because we almost have a new company now. We've never seen a normal mortgage market with our current structure. So again, last year we had 11% margins, which are close to peak margins in a market that was roughly as size of the peak market, right. So our goal is to always run in double-digit margins. We always want to run at double-digits margins. I think next year, that it will be under pressure in '14, just because refis are falling off. But we feel like '14 will be the trough in terms of earnings, and '15, '16 and beyond we feel like there's long-term growth in our business.
It's hard to say what peak could be, but at least double-digit margins. And I think that it depends on what you're think a normalized market is, but if we were in one-two to one-five market, with two-thirds purchase and one-third refi, we could have meaningfully higher than 10% margins.
You talked about the cost that you worked out structurally, I'm just curious if you can quantify what those might have been last year, had you not made the changes that you've done?
I don't have a specific number. We did do some assets like a year or two ago and we took out $1billion of cost out of the business, $1 billion from the peak. Now that was back when our company was $6.5 billion to $7 billion of revenue, and now we're at, call it, $4.5 billion of revenue.
So $1 billion of those costs were real hard costs, like people and systems that we took out and things like that. I don't really have a number for like what it was in last year, but I would just say it was meaningful. I don't know if I can give you more details on that.
No. The CoreLogic that was all kind of a title company, so that's just $1 billion of personnel cost and other operating expenses. Not claims, not taxes. That's just real hard cost from the peak.
So on your slide showing the 11% or the high margins, that's relative to history. In concerning that commercial as a higher margin product than retail and then fourth quarter last year was a great commercial quarter. Is there a mix change here that we're seeing that is more commercial versus residential that's driving that margin back up to peak levels.
We are getting the benefit of commercial, but I would say that that's not really driving the 11% margin. That might be a little bit of it. I think in 2012 what happened was we had several years, where the market was just continuing to fall and yet we were still cutting our structural costs. And finally in 2012, refi spiked up quickly. And we saw the earnings power that First American actually have based on our new cost structure. So I think the biggest thing was, it was probably more refi driven and commercial driven. And it was very much driven by our new cost structure.
And one thing I would add is, 2012 was the first year since 2006 that we saw revenue growth that revenue up 18%, and we did a fantastic job of leveraging that to the bottom line.
From kind of a longer-term industry perspective, I guess, you have a lot of agents that are sharing and a lot of the revenue is getting generated off of your underwritings. Is there a longer-term ability to kind of disintermediate from that or centralized a larger share of originations over time or you just think it's a permanent feature of this market?
I think over time the shift will slowly drift more toward direct, slowly, given new regulations that the CFPB has taken out, given the tougher standard that the lenders are telling [indiscernible] agent, so they have to abide by. And a lot of these smaller agents they just don't have the capital requirements, they don't have the systems, the infrastructure, and a lot of lenders are acquiring now, free to close the loan with Wells Fargo or Bank of America.
So I think it will slowly drift more towards the direct side. That being said, there are certain stage that there are just always agent states, Massachusetts and a lot of New England states, it's always agent, because you need a lawyer to close the transaction at many places.
I think we have to conclude there. They will be available for break out in the Morgan Suite afterwards. Please join me in thanking him for their time.