Prepared remarks
Greetings, and welcome to the First American Financial Corporation Second Quarter Earnings Conference Call. At this time, all participants are in a listen-only mode. A brief question-and-answer session will follow the formal presentation. A copy of today's press release is available on First American's website at www.firstam.com/investor. Please note that the call is being recorded and will be available for replay from the company's investor website and for a short time by dialing the replay numbers listed on the investor website and entering the conference ID provided on that site. We will now turn the call over to Craig Barberio, Vice President, Investor Relations, to make an introductory statement.
Good morning, everyone, and welcome to First American's Earnings Conference Call for the Second Quarter of 2026. Joining us today on the call will be our Chief Executive Officer, Mark Edward Seaton; and Matthew Feivish Wajner, Chief Financial Officer. Some of the statements made today may contain forward-looking statements that do not relate strictly to historical or current fact. These forward-looking statements speak only as of the date they are made and the company does not undertake to update forward-looking statements to reflect circumstances or events that occur after the date the forward-looking statements are made. Risks and uncertainties exist that may cause results to differ materially from those set forth in these forward-looking statements. For more information on these risks and uncertainties, please refer to yesterday's earnings release and the risk factors discussed in our Form 10-Ks and subsequent SEC filings. Our presentation today contains certain non-GAAP financial measures that we believe provide additional insight into the operational efficiency and performance of the company relative to earlier periods and relative to the company's competitors. For more details on these non-GAAP financial measures, including presentation with and reconciliation to the most directly comparable GAAP financials, please refer to yesterday's earnings release, which is available on our website at www.firstam.com. I will now turn the call over to Mark Edward Seaton.
Thank you, Craig. Our earnings momentum continued in the second quarter as we generated adjusted earnings per share of $2.80, an increase of 36% from the prior year. Commercial continued to be a standout performer. Revenue increased 34%, setting a second quarter record. We closed 14 transactions, generating more than $1 million of premium, up from 11 a year ago. Within our National Commercial Services Division, demand remains broad based with 10 of 11 asset classes growing year over year. Purchase revenue increased 2% as affordability challenges continue to weigh on existing home sales. Refinance revenue increased 18%, reflecting the brief surge in open orders we experienced at the end of the first quarter when mortgage rates reached their lowest level since 2022. While that activity provided a tailwind during the second quarter, volumes have moderated as mortgage rates have moved higher again. One of the most important earnings drivers continues to be our bank, First American Trust, which provides a growing source of investment income. During the quarter, average deposits totaled $7.9 billion, an increase of 30% from last year. Growth was driven by deposits outside of our captive title business. During the quarter, 36% of deposits came from sources beyond our captive title operations. The largest contributor was ServiceMAC, our mortgage subservicer, which accounted for $1.7 billion of deposits, up 76% from last year. ServiceMAC's loan portfolio grew 54% during the quarter, and as that portfolio expands, so should its deposits. Our second largest source of non-title deposits came from our 1031 Exchange business. Last year, all exchange deposits were held at third-party banks. Since launching our 1031 banking solution less than one year ago, we have rapidly grown deposits, which averaged $827 million in the second quarter, representing roughly one-third of our total 1031 balances. Finally, our agent banking strategy continues to gain traction. Today, 10 title agents bank with First American Trust, an increase of 37% from last year. We expect those balances to grow as real estate activity recovers. Taken together, servicing 1031 Exchange and agent banking provide meaningful long-term growth opportunities while reinforcing the bank's role as a valuable countercyclical earnings driver. Our primary strategic priority remains leveraging AI across the enterprise to amplify the talents of our people, better serve our customers, and strengthen our operating capabilities. These benefits are already becoming tangible. Recently, we needed to update 1,300 forms across the company. Historically, this would have required a lengthy manual process. Using our new AI tools, we reduced the time required by 97%. We launched a product called ExamAssist QC, an AI-enabled quality control workflow. It has now processed more than 50,000 orders, delivering 92% with no additional human review. A clear example of how we can deploy AI at scale for a quality control process. We are also starting to see meaningful evidence that AI can improve customer-facing service delivery. At ServiceMAC, we rolled out a virtual agent last month for loan inquiries and improved self-service success from 0% in April to 42% in June. While still early, it is a useful proof point that AI can support live customer workflows in a regulated servicing environment. We expect to expand the number of self-service use cases from one to seven by the end of the year. We are also building broader enterprise capability in agentic product development. In the past four months, we have had nearly 700 people participate in hands-on boot camps focused on rewriting legacy code and solving real business problems. The result is a growing enterprise capability to apply agentic AI across functions and workflows, moving technology teams from basic awareness to real adoption in product development. And of course, at the enterprise level, we are fundamentally reimagining title and settlement through Endpoint and Sequoia. Both platforms continue to achieve important milestones. Beginning with Endpoint, we remain on track to scale the platform across our local title branch network by the end of 2027. During the quarter, we converted our First American Title office in Spokane, Washington. While it is still early, every indication suggests the transition has been successful. Escrow professionals now operate from a platform where agentic AI automates routine tasks, which will allow our teams to spend more time serving customers and managing complex transactions. This quarter, we will expand Endpoint across additional offices in Western Washington before completing a statewide rollout by year end, followed by a broader national deployment throughout 2027. We have also improved automation rates from 30% in Q1 to 34% in Q2, and so far in July, we were at 39%. We expect those rates to improve as the platform matures. This represents a fundamental shift in how title and settlement work gets done. As workflows become standardized, the role of our people increasingly shifts from executing routine tasks to validating AI-generated work and focusing on higher-value customer interactions. We also continue to make excellent progress with Sequoia, our AI-powered title decisioning platform. Since our last earnings call, we expanded Sequoia's refinance capabilities beyond our local direct operations into our centralized lender division in Southern California. We also broadened our refinance coverage in California, increasing our footprint from 8 counties to 41. During the quarter, our automation rate improved from 35% to 40%, and we expect further gains as the platform continues to learn and mature. Purchase transactions remain a more complex challenge. We launched purchase capability in three counties during the first quarter and expanded into Orange and San Diego Counties during the second quarter. Currently, in these counties, Sequoia provides instant title decisioning for approximately 16% of purchase transactions at order opening. Over time, we believe we can automate title decisions for approximately 70% of purchase transactions and 80% of refinance transactions in markets where we maintain title plants. That capability is made possible by our industry-leading title plant data, deep underwriting expertise, and innovative technology. By year end, we expect Sequoia to be deployed across California and Florida, with a broader national rollout plan for 2027. Once Endpoint and Sequoia are fully rolled out, we believe they will create a durable competitive advantage, improving the experience for employees, delivering better service for our customers, and creating meaningful long-term value for shareholders. Turning to our outlook, we remain optimistic about our earnings for the second half of the year. Six months ago, we said our commercial business was on pace to deliver a record year, and we continue to believe that. Our commercial pipeline has never been stronger. We have already closed three transactions generating more than $1 million in premium during July, and commercial opened orders were up 9% over the first three weeks of the month. We remain more cautious than the broader consensus on the residential purchase market. Through the first three weeks of July, our open purchase orders are flat relative to last year as existing home sales remain sluggish. Finally, I will comment on capital management. Our business continues to generate substantial and growing cash flow. During the first six months of the year, our free cash flow was $285 million, up 32% relative to last year. This is a result of improving operating cash flow and declining capital expenditures which were down 18% year over year. We expect cash generation to strengthen during the second half particularly since the first quarter is our seasonally weakest period. Our first capital allocation priority remains investing in the technology, platform and products that will extend our leadership position in the industry. Importantly, these investments are already embedded within our existing run rate. In fact, our companywide technology spend has remained relatively flat since 2022, and we do not anticipate the need to invest materially more in our business than what we are currently investing. Our second priority is acquisitions. The bar for acquisitions is higher today than it has been in many years. We are pleased with our geographic footprint and portfolio of businesses and we have no interest in pursuing acquisitions simply for the sake of scale or diversification. However, we will continue to pursue opportunities that have strong strategic synergies with our current business, whether in title or near adjacencies. Finally, we remain committed to returning capital to shareholders through a combination of dividends and opportunistic share repurchases. Expect us to continue increasing our dividend over time, reflecting our confidence in the company's long-term earnings growth. We will also repurchase shares when we see attractive opportunities like we did in the second quarter. In summary, we remain intensely focused on reimagining title and settlement through AI. We have a strong balance sheet and disciplined strategy, unique assets like First American Trust, and industry-leading title data that position us to capitalize on the transformational opportunities AI presents. Together, these strengths give us a differentiated competitive advantage and position us well for years to come. Now I will turn the call over to Matthew, who will discuss our financial results in greater detail.
Thank you, Mark. This quarter, we generated GAAP earnings of $2.12 per diluted share. Our adjusted earnings, which exclude the impact of net investment gains and purchase-related intangible amortization, were $2.08 per diluted share. Focusing on the title segment, adjusted total revenue was $2.0 billion, up 14% compared with the same quarter of 2025. Commercial revenue was $314 million, a 34% increase over last year, driven by a 31% increase in average revenue per order. Average revenue per order was $20,000 per transaction, which reflects a record level for our commercial business. Purchase revenue was up 2% during the quarter due to a 6% increase in average revenue per order, partially offset by a 3% decline in closed orders, which reflects the continued weakness in home sale activity. Refinance revenue was up 18% compared with last year, due to a 12% increase in closed orders and a 5% increase in the average revenue per order. This growth was supported by a temporary decline in mortgage rates earlier this year, though activity has since softened as rates have moved higher. Refinance accounted for just 5% of our direct revenue this quarter and highlights how challenged this market continues to be compared to historic levels. In the agency business, revenue was $820 million, up 14% from last year. Given the reporting lag in agent revenues of approximately one quarter, these results primarily reflect remittances related to first quarter economic activity. Information and other revenues were $295 million during the quarter, up 12% compared with last year. The increase was driven by revenue growth at ServiceMAC, higher demand for non-insured information products and services, and refinance activity in the company's Canadian operations. Investment income was $164 million in the second quarter, up 11% compared with the same quarter last year. Despite the Fed cutting rates three times, the increase was primarily due to higher interest income from the company's investment portfolio driven by growth in the size of the portfolio. The growth in the portfolio was attributable to the increase in deposit balances at First American Trust that Mark discussed. Personnel costs were $572 million in the second quarter, up 9% compared with the same quarter of 2025. The increase was mainly due to incentive compensation expense resulting from improved financial performance and higher salary expense. Other operating expenses were $319 million in the quarter, up 15% compared with last year, primarily attributable to higher production expense driven by higher volumes and increased software expense. Our success ratio for the quarter was 66%. This is somewhat higher than our target of 60%, primarily due to investments in certain businesses outside of our domestic title operations such as ServiceMAC. The investments being made at ServiceMAC are to support the meaningful growth in its loan portfolio. The provision for policy losses and other claims was $45 million in the second quarter, or 3.0% of title premiums and escrow fees, unchanged from the prior year. The second quarter rate reflects an ultimate loss rate of 3.75% for the current policy year and a net decrease of $11 million in the loss reserve estimate for prior policy years. Interest expense was $30 million in the current quarter, up 33% compared with last year due to higher interest expense related to the growth in deposit balances at First American Trust. Pretax margin for the title segment was 15.7% or 14.0% on an adjusted basis. Moving to the home warranty segment, adjusted total revenue was $112 million this quarter, up 1% compared with last year. The loss ratio was 40%, down from 41 in the second quarter of 2025. The slight improvement in the loss ratio was due to lower claim frequency partially offset by higher claim severity. Pretax margin in the home warranty segment was 21.3%, or 20.2% on an adjusted basis. The effective tax rate in the quarter was 22.8%, which is slightly below the company's normalized tax rate of 24%. Our debt-to-capital ratio was 31.4%. Excluding secured financings payable, our debt-to-capital ratio was 21.5%. During the quarter, we repurchased 330,000 shares for a total of $20 million at an average price of $61.99. Now I would like to turn the call over to the operator to take your questions.
Questions and answers
Thank you. We will now be conducting a question-and-answer session. If you would like to ask a question, please press 1 on your telephone keypad. You may press 2 if you would like to remove your question from the queue before pressing the star keys. Our first question will come from Terry Ma with Barclays.
Hey. Thank you. Good morning. Maybe just on the deposit growth, can you expand on some of the comments and maybe for the ServiceMAC piece, how sustainable is that above-average kind of deposit inflow? And as we look to the back half of the year, what is the cadence of investment income?
Thanks for the questions, Terry. I will start with the deposits, and Matthew can talk about investment income for the back half. We have a bank, and it is a real strategic advantage for us. For many years, we put our own First American Title deposits that we manage in connection with the escrow process into our bank, and we have really maximized that. About five years ago, we decided instead of just providing banking services to our own First American Title Insurance Company, let's provide banking services to others within the title industry. There are a lot of agents out there that manage escrow deposits and they put their deposits at third-party banks. These are customers of ours. We started off with agent banking, and we are making progress on that, as I talked about. There are about 20,000 different settlement agents out there; not all of them are going to want to use First American Trust, but many of them will. We are making really good traction there, and it ties our agents closer to us, which is a good thing. In the meantime, over the last couple of years, we found other sources of deposits, including the 1031 solution and ServiceMAC. ServiceMAC is growing despite a flat market; their loan growth is up 54% from last year. Those deposits tend to come to us when customers are indifferent or where we can capture them, and we try to push those to First American Trust whenever possible. So we feel like it is sustainable in terms of where we are with these third-party deposits. With that, I'll turn it over to Matthew to talk about investment income.
Yeah. Thanks, Mark. Hi, Terry. So investment income, like I discussed, was up 11% year over year, driven by growth in the investment portfolio related to the increase in deposits at the bank. At the same time, interest expense also grew year over year; interest expense grew 33% year over year. When I look at investment income, I like to look at it net of interest expense. Investment income net of interest expense grew 8% year over year, and I think that 8% is a good proxy for the growth that you will see in the back half of the year.
Got it. That is helpful color. And then as my follow-up, maybe just on the commercial ARPO — it continued to see robust year-over-year increases. Certainly appreciate all the color on the larger $1 million-plus premium deals. What is the outlook for that in the second half? And ultimately, how sustainable are those ARPO increases as we look out to the back half of the year? Thank you.
Yes. Thanks a lot, Terry. On commercial, we are very bullish. Our order counts continue to grow, and our fee per file or ARPO continues to grow. We are getting a lot of bigger deals now. The big-deal pipeline is really strong. We think ARPO will continue to grow in the second half of the year. One of the things we get from investors a lot is how sustainable is this commercial market — is this going to go away? We feel like the commercial market has legs for a lot of different reasons. When we look at our pipeline, we have conversations with our customers and we analyze the commercial real estate dynamics. We are still in the early innings of the next commercial real estate cycle. So we feel really good about commercial and our ARPO for the second half of the year and well into next year. Thank you.
Thanks, Mark.
Our next question will come from Oscar Nieves with Stephens.
Hey, good morning. My first question is on margins in the title segment, which were strong at 14% — that is roughly an 80-basis-point expansion year over year. Can you give us a sense of where you see the full-year margin landing at this point, and whether the back half plays out differently than the first half given the comps?
Hi, Oscar. This is Matthew. Year to date, our margin in the title segment is 12.3%. When we look at the back half of the year, I think we can expand on that, but the level of expansion that we get from the 12.3% is going to be tied closely to the commercial business, which, as you know, is hard to forecast, particularly the strength of it in Q4.
Okay. That helps. Related to margins, when we look at the trends in operating expenses, your personnel and other OpEx ratio improved nicely year over year. If we look at the incremental margin this quarter specifically, it looks like it was a little less efficient than what you posted during the first half overall. Another way to say that, if you look at the success ratio, the rate is a little bit mixed there. What can you share with us on that?
Thanks, Oscar. The way we look at efficiency is the success ratio, which is the change in net operating revenue divided by the change in personnel and operating expenses. Our target success ratio is 60%. That remains a good target for our business, although it can change quarter to quarter based on one-time items or investments. In Q2, the success ratio was 66%, a little above our target. That was due to investments in businesses outside of our domestic title operations, such as ServiceMAC, where we are investing to support significant growth in their loan portfolio. Looking ahead, we may see the success ratio a bit elevated relative to our 60% target due to these investments, and one-time items can also impact the ratio, particularly as we compare to Q4 2025, which had some one-time items benefiting the title segment.
Super helpful. And just one last one on capital allocation, specifically on buybacks. You bought back about $20 million of stock in 2Q. How are you thinking about the pace of buybacks from here to the balance of the year? Has the recent increase in your debt-to-capital ratio changed that thought process at all?
On buybacks, we are not in the market at the moment, but it is something we look at constantly. Over most of the last five years, in most quarters we have been repurchasing shares. We look at dislocations in the market as opportunistic moments to buy shares, and we think buybacks have been good for our shareholders at the prices we've paid. The debt-to-capital ratio does not really change that decision now. Our target debt-to-capital is 20%, and we're a little higher than that now, but still very comfortable, especially considering we're kind of in a trough in the market. I do not think the current debt-to-capital level weighs on the buyback decision.
Thank you. That is all I have. Thank you so much.
We will go next to Bose George with KBW.
Hey, guys. Good morning. The 6% you noted on purchase ARPO seems a lot higher than sort of home price appreciation itself would imply. Is there more activity at the higher end of the market, or any color to add on that?
Hi, Bose. It's really due to mix, particularly California. We had a higher mix of orders coming from California, which has a higher average revenue per order.
Okay. Great. Makes sense. Thanks. And then on the commercial side, can you remind us what the biggest buckets are? How much of the premium is coming from data centers and energy? Are those two the biggest buckets?
We track 11 asset classes. Our biggest asset class is industrial — it represented 23% of our premium. Some data center work is included in that bucket, but industrial also includes warehouses and other facilities. Multifamily was 16% of our premium. Development sites were 14% of our premium; data centers can be included in development if it's raw land that will be built into a data center. Retail was 14%. Those are our top four asset classes.
Okay. Great. And then actually one more on the political front. In late June, Bill Pulte posted a comment on social media about FHFA working on expanding title and that something could be expected soon from Fannie Mae. Have you heard anything incremental about that?
We have not heard anything incremental. We know they've extended the title acceptance pilot through November 2027, and beyond that we have not received additional information relative to the comment you mentioned. So we are in a wait-and-see mode.
Okay. Great. Thanks a lot.
We will go next to Mark DeVries with Deutsche Bank.
Yeah. Thanks. I have some follow-ups on commercial. You said earlier that you are seeing strong growth across 10 of the 11 different asset classes. Can you talk about where you are seeing the strongest growth across those asset classes with a particular focus on data centers and office?
Give me a second here. Our development site bucket is up 33% from last year. Multifamily has grown 23% from last year. Retail is up 59% from last year. When you look at everything except data centers, our commercial business is up 11%. Data center revenue is up substantially — we saw very strong growth in that area, but the point is we are seeing broad-based growth across many asset classes, which gives us confidence that this market has legs.
Is office the one that is not growing? Are you seeing any green shoots there?
We have not seen much in terms of office. Of our 11 asset classes, the only one that is not growing year over year is health care, so overall commercial is broadly growing, but office remains challenged relative to other asset classes.
Okay. Got it. And then turning to data centers, could you help us think about how premiums on that compare to the average commercial transaction? Also how the premium size differs across the three discrete revenue opportunities you get with the average data center transaction?
With data centers, typically the principal will buy land — that's one transaction. They will get a construction loan to build the data center — that's the second transaction. And there is a takeout refinance — the third. Many of these deals are very large; some are billion-dollar deals. When you look at ARPO growth, much of it is driven by these huge deals. The average ARPO for a data center deal is well above our overall average ARPO. Many of these deals generate million-dollar-plus premiums.
So it is not that there are many of them, but the ones you get have very high premiums. Of those three premiums you receive, is the land the smallest and each one progressively larger? How does that typically rank?
I am not sure I can rank them precisely off the top of my head. Typically, the takeout refinance at the end is probably the smallest premium, but the first two can vary and I'd have to do some analysis to rank them definitively.
Do you also in the policy insure the actual servers and equipment in the building, or is it just the building itself?
When a principal gets a title policy, they typically get it for the amount that it takes to build the data center, which can include equipment. If the data center fails to operate because of servers or equipment issues, those are not typically title-related risks that we are on the hook for, but the insured amount can reflect the full project cost, including equipment.
Okay. Got it. Thank you. Yep. Thanks a lot, Mark.
There are no additional questions at this time. That concludes this morning's call. We would like to remind listeners that today's call will be available for replay on the company's website for a short period. The company would like to thank you for your participation. This concludes today's teleconference. You may now disconnect.