Graham Holdings Co. Stock

All right. It seems like we're getting our last cups of coffee here. So we'll get going. I'm just going to start with a brief legal disclaimer. So we're all aware. So a full disclaimer is posted on the screen behind me and the company's website, along with the meeting presentations. The information presented at this meeting may contain forward-looking statements based on the company's current expectations. Forward-looking statements are subject to various risks that could cause actual results to differ materially from those stated. These statements should be considered in conjunction with the risks and uncertainties described in our filings with the SEC, including our most recently filed reports of Form 10-K and Form 10-Q and subsequent filings.

In addition to the results reported in accordance with you, generally accepted accounting principles included in this presentation, the company is providing certain non-GAAP financial measures. A reconciliation of such non-GAAP financial measures to the most directly comparable GAAP financial measures can be found in the appendix of the presentation on the company's website. Also going to start giving the last 5 seconds of a commercial where they go and report everything.

Okay. Well, welcome to everybody to our 2026 Graham Holdings Annual Meeting. I thought I would -- I would start with the formal proceedings, and then we'll go to some management commentary from there. So good morning, ladies and gentlemen. The meeting will please come to order. I'm Tim O'Shaughnessy, President and CEO of Graham Holdings Company. I will act as Chairman of Meeting. To my right is Wally Cooney, SVP and CFO of the company; and Nicole Maddrey, our SVP and General Counsel and Secretary of the company, will act as the Secretary of the meeting. And I'd like to welcome you all to the Annual Meeting of Stockholders for 2026.

Now I'll briefly describe what is on the program this morning. So first, we will dispense with the technical part of the meeting, which involves such matters as the submission of documents and the determination of a quorum. After my remarks, we'll hear from Andy Rosen, who will provide an update on Kaplan. Then we will proceed to the election of directors and the proposal to be voted on by the Class A shareholders to approve the 2025 compensation awarded to the named executive officers. After that, the meeting will be open for your comments and questions.

So before turning to the opening formalities of the meeting, I would like to introduce those nominees for election as a director who are present: Don Graham, Anne Mulcahy, Tony Allen, Danielle Conley, Chris Davis, Tom Gayner, Jack Markell, Rick Wagoner, Katharine Weymouth and myself. Also present from PricewaterhouseCoopers, the company's independent registered public accounting firm, are Rob Vasco and Celia Jim.

Will the Secretary please present the meeting all the supported documents?

For the purposes of this meeting, I present affidavits of mailing of the notice of availability of proxy materials for the 2026 Annual Meeting of Stockholders to each stockholder of record at the close of business on March 11, 2026, the record date for determining stockholders entitled to receive notice of and to vote at this meeting. The complete list of the holders of Class A and Class B common stock as of the close of business on March 11, which has been available for at least 10 days preceding this meeting. A copy of the certificate of incorporation and the bylaws of the company and the minutes of the last annual meeting of stockholders of the company held on May 6, 2025.

Alisa Zagare and Elaine Wolff have been appointed to act as inspectors of votes at this meeting. I direct that an executed copy of their oath be filed with the records of the meeting.

Will the Secretary please ascertain that a quorum is present?

Mr. O'Shaughnessy, the inspectors of votes have canvassed the stockholders present in person or by proxy and have presented to me their first report, which shows that there are present in person or by proxy 27 stockholders holding 928,001 shares of Class A common stock of the company, which is 96.27% of the Class A common stock entitled to vote at this meeting and not less than 2,625,155 shares of Class B common stock of the company or 78% of the 3,385,088 shares entitled to vote at this meeting.

I direct that the first report of the inspectors of votes be filed with the records of the meeting and declare that a quorum is present at this meeting may proceed to the transaction of the business for which it has been called.

As stated in the notice of the meeting, the purposes of the meeting are to elect the directors of the company with Class A shareholders on an advisory basis, to vote to approve the 2025 compensation awarded to named executive officers and to transact such other business as may properly come before the meeting or any adjournments thereof.

Okay. I thought I would begin. Many of you may have seen that I'm walking a little different than normal. And so for those that haven't asked and well, after I'll get this out of the way. I crashed a mountain bike with my son and I have an injured foot, which is better than the alternative, which was I presented the operating projections to Don, and I came out of the meeting with this. So I'm glad it was the mountain biking injury. But -- so good morning, and once again, and welcome to the Annual Meeting of Shareholders of Graham Holdings. We're delighted to have many of you in person and also on the audio stream. There are a bunch of members of management here as well. So hopefully, you've been able to introduce yourself and they'll stick around for a bit after if you would like to go and chat at all.

Our agenda for the content portion of the meeting will be as follows. I'll provide a brief update on operations for 2025 and Q1 of 2026 as well as discuss the company's approach to AI. Andy Rosen will update you on Kaplan's operations with a focus on Kaplan International. And then we'll open up the floor for questions for as long as time allows.

So we were pleased with 2025 results and feel similarly about Q1 2026. For the quarter, total company revenue grew 6% over the prior year, with large gains in health care and manufacturing being modestly offset by declines in automotive and other businesses. Adjusted operating cash flow also grew in Q1 to $113 million, up 28% from the prior year. Improvements at education, broadcasting, manufacturing and other businesses were partially offset by declines in health care and automotive. With CapEx running slightly higher in Q1 of this year, adjusted free cash flow was just below that of adjusted operating cash flow coming in at 24% growth over the prior year.

So zooming out, the revenue of the company has grown pretty respectively with an 11.4% CAGR over the past 5 years. Newer segments such as automotive and health care have become larger pieces of the pie, helping to drive that growth. But more importantly, adjusted operating cash flow grew at an 11.6% CAGR, in line with that revenue growth. This drove the business from $253 million in adjusted operating cash flow in 2021 to $407 million in 2025.

We have occasionally presented the business with Graham Media Group and our corporate office expenses excluded in order to give a better sense of the underlying growth occurring at many segments of the company. We do so again here, although I suspect this might be the last time I do as the relevance of this is meaning, growth in the rest of the company has made the comparison less useful over time as the results are now more clearly represented in the consolidated numbers. But what this comparison shows is that many segments of the company have been growing substantially.

Over the last 5 years, the operating businesses, excluding Graham Media Group grew adjusted operating cash flow at a 24.1% CAGR from $142 million in 2021 to $337 million in 2025. In absolute terms, this growth was led by Kaplan, which increased by $102 million to $213 million in 2025. However, Kaplan actually went down as a percentage of overall adjusted operating cash flow from this group. 2021, Kaplan represented 78% of the non-GMG operating cash flow. By 2025, this number has decreased to 63%.

We like to see this as it implies we have multiple segments that have generated outstanding growth over a sustained period, let's dive in the operating segments more fully. Andy will discuss Kaplan extensively in just a few minutes, so I'll keep my commentary brief. Q1 revenues were up 4% and adjusted operating cash flow increased by 15% to [ $62 million ]. The business is performing well through tricky currents, but most intensely at Kaplan International. The team is navigating the challenge as well. You like me, how much to be proud of with Kaplan's performance on all measures.

It also feels like a good time to touch on the recently closed transaction to sell Kaplan Languages Group. Andy will discuss how this fits into our larger strategy at Kaplan, but I wanted to share a few vital statistics. The transaction closed this past May 1, consideration is tied to future performance, which, if [ MAX ] will not be material to Graham Holdings, we expect to record a current U.S. income tax benefit of approximately $60 million in 2026, and the brand will transition away from Kaplan over the course of 2026. I'd like to thank everyone associated with successfully managing and completing this transaction.

The Broadcasting segment is another sector that hasn't been quiet as of late. I'll start by saying what you might expect. We are closely following the Tegna and Nexstar merger and the implications for the space, depending on the ultimate outcome. This does not directly impact our business today. We will continue to manage the business well, look for opportunities to strengthen our operations and continue to search for new revenue.

As for operations, Q1 was a strong quarter with revenue growing by 8% and adjusted operating cash flow by 31% to $39 million. Results improved due to the Super Bowl and the Olympics' presence on our NBC affiliates as well as political advertising and primary campaigns. We continue to expect 2026 to be a robust year, due in large part to the Senate campaign in Texas and the Gubernatorial and Senate races in Michigan, which would drive significant political advertising in our markets.

Catherine Badalamente and her team have also worked in the last year to bring about rebuilt newsroom processes that should, over time, allow us to increase the percentage of our resources that go into news gathering and reporting. This is an ongoing effort to allow us to adapt from legacy processes to continue to best serve the community and report the news in today's day and age.

Our Healthcare division saw robust growth in 2025 with adjusted operating cash flow reaching $115 million, surpassing the $100 million mark for the first time. Growth was led by CSI Pharmacy, which continues to scale this footprint and offerings in the in-home infusion market. Revenue growth continued into Q1, with CSI growing 31% in the overall division growing 20%. Adjusted operating cash flow declined for the quarter by 8% to $21 million, with CSI down $3 million from the prior year.

Some understandably might find it surprising to see operating income declines with such robust revenue growth. This decline was driven by several factors, notably an investment in a new long-term incentive compensation plan, modest gross margin declines due to pricing and unfavorable mix shift and most notably, investment in infrastructure for future growth.

Many of you know and hopefully view as a feature of Graham Holdings that we do not manage the business for any particular quarter. We think there's a good opportunity to grow the long-term value of the company, expand our moats and increase the intrinsic value of business for shareholders. We will do so regardless of the impact on a particular quarter. The result, as we may occasionally have depressed earnings in the name of a greater good. Q1 is an example of this approach. We are not running CSI for maximum 2026 cash flow and have been willing to slow earnings growth in the name of future cash flows.

You may recall, several years ago, I indicated we were investing in CSI and hopes of it becoming a bigger business. This belief came true. For a few quarters, revenue grew with minimal flow through to the bottom line, but subsequent -- by the subsequent few years to rapidly climbing income. It's been about 3 years since we undertook that cycle and the business is once again ready to take steps in order to position itself to become a larger business with greater earnings growth. We're optimistic we will be successful this time around as well.

While this isn't comprehensive, I'll provide you with some of the major expenditures that materialized in Q1 to help provide a sense of what we are building at CSI. We opened a new pharmacy location in Denver that began ramping in earnest in Q1. This will allow for greater coverage in the Mountain West and Pacific Northwest. In Q4 2025 and Q1 2026, we hired and trained a large number of new salespeople as we began expanding into biologics and are attempting to further increase our market penetration in IVIG. The sales team increased in count by 34% from the end of Q3 2025 through Q1 2026, with most of the new team members still in either training or ramp phases in Q1.

We added several new strategic roles to help manage our manufacturer and payer relationships as well as established a new long-term incentive plan that rewards management for profitably growing business. This plan began accruing expenses in Q1 of 2026.

Over an extended period of time, we expect CSI's cash flow generation to align with revenue growth, but it won't always be a smooth line. The opportunity set will dictate the pace and trajectory of growth. We do, however, expect these investments to be less pronounced in the P&L as the year progresses.

At Graham Healthcare Group, our other scaled business within the division, results were good and in line with expectations. Dee Grein is off to a very good start as CEO and in fact, completed her first acquisition with the purchase of Covenant Home Health based in Eastern Pennsylvania. This transaction, while relatively small, something we'd love to replicate. Covenant has strong operations in the Philadelphia home health market. This provided adjacent geographic expansion to our existing Pennsylvania business and further grew our market share and importance to Pennsylvania-based payers.

Additionally, when put into the GHG operating structure, it will become a better business economically. That integration is underway and Dee and the team are optimistic about the future of covenant and hopeful about more opportunities like it.

Manufacturing segment saw strong year-over-year growth to kick off 2026 with revenue increasing 28% and adjusted operating cash flow increasing 38%. This growth was led by Hoover, primarily due to the inclusion of our recently acquired aluminum cladding operation as well as growth at Joyce, our linear motion business. Recent trends remain strong, and we are optimistic about what the remainder of 2026 will bring within this segment.

Our tour around the world continues with our automotive group. Results were down in Q1, both at the revenue and operating income levels. Our exposure to the DC Metro area has been a drag for the business in recent months. Q1 results were significantly impacted by reduced DC area economic activity as well as the snowcrete weather event that would severely impacted both sales and repair order volume for several weeks. While we do not expect the latter event to repeat, we are watching DC area economic activity closely.

Lastly, our Roda service business continues to gain strong traction. Differentiated experience is built with the customer in mind from the ground up. Our customer satisfaction and NPS scores continue to be best in class for the industry. We expect growth to accelerate later this year when our new Virginia-based facility opens, allowing for significant expansion in our service area.

Out last stop is the other businesses segment. Results were mixed with revenue decreasing modestly, but adjusted operating cash flow improving by 21%. We expect continued improvements in adjusted operating cash flow as the year progresses. As a note, revenues, excluding World of Good Brands, which we sold in 2025 were up modestly.

Within the segment, Framebridge continues to see strong growth as its retail expansion quickens. Year-to-date, 3 new stores have opened with the pace expected to accelerate in Q3. Clyde's Restaurant Group performed admirably through the previously mentioned snowcrete weather event as well as a massive water main break directly in front of our building that closed our Georgetown location for several weeks. If you've never seen 4 feet of water in a basement, I can show you pictures.

Lastly, our digital media operations continue to feel the impacts of reduced search referral traffic in the age of AI results and engines. While our content continues to be very valuable to LLM, quid-pro-quo of traffic in exchange for content has not yet been reestablished.

Let's take a step back to some corporate level observations. Our balance sheet ended the quarter with a $344 million surplus of cash and securities balance as compared to debt. Our securities balance declined in Q1. We were a net seller of securities in addition to price declines.

We also began to repurchase shares in Q1, buying 32,190 shares at an average price of approximately $1,061 per share. We've continued repurchasing since the quarter's end. Thought it might be useful to reiterate our philosophy on share repurchases. We only buy when we think there is a meaningful discount to intrinsic value, conservatively calculated. Price always matters. We do not have a set it and forget it program for repurchases. We will never repurchase shares because we think it will cause a short-term boost in the stock price. We will not repurchase shares if we think it could introduce risk to the balance sheet or limit our ability to act opportunistically elsewhere.

So I wanted to spend a little bit of time talking about our approach on AI here as well. And as discussed in the annual letter, we're focused on AI and how it will impact Graham Holdings. Several of those paragraphs summed up my view very well, which I want to reiterate here.

To be blunt, we think AI technologies represent a shift in how business will operate. We do not plan to understand the long-term impacts on how employment will change. We do believe that employees and organizations that effectively harness technological shifts and AI tools will be substantially advantaged as compared to those that do not. We are spending much time at our businesses, understanding how AI will change our operations. We are optimistic.

Some of the things we think we think. Full Stack employees and leaders will have the most success in our organizations. People who can build applications, natively use AI tools and understand how it drives the business model will drive the most success and have the most personal success. If you are not vibe coding with AI tools, you'll be at a structural disadvantage. People and process-heavy tasks are where we're most likely to see immediate improvements.

In many industries, we think it's a watershed moment. There will be those who adopt AI tools to drive improved business outcomes in a timely manner and those who do not. Improvements in business model will drive market share gains for the house, leaving the have nots to become fundamentally impaired. Our operations should be able to drive the cost of leverage in general and administrative functions, but in most areas in our companies have room to improve.

Lastly, product development time lines at most of our businesses should be measured fractionally as compared to the past.

In 2025, we hired a Chief Technology Officer, Spiro Roiniotis in the Graham Holdings corporate office. He has worked at several of our businesses dating back to 2009. Spiro and his small team have a challenging job, and our expectations are high. They will be working with many of our businesses to accelerate their understanding and implementation of AI tools to drive real product improvements and improved economics.

Lightweight engagement with AI is not a realistic option for organizations that want to thrive in the next decade. You can rest assured that Graham Holdings is setting the cultural tone that provides the best chance of success in our businesses. We believe that this is not outsourceable. All levels of the organization need to and must engage from the CEO to the most junior employees. Yes, I am building apps and skills and have only half-jokingly mentioned, it might be my highest and best use to the organization.

If we get this right, we'll build a stronger, more valuable company, but the acceleration of the AI era brings about an interesting set of questions to a holding company model. How can we help our management teams that come from very different backgrounds and experiences prepare to engage aggressively to understand the new technologies and how their industries may evolve. As the game changed entirely for any of our companies. If so, how or will, core strategies need to change. Do we need to re-underwrite our perspectives on the possible and likely economic outcomes for any of our businesses.

I wanted to also briefly cover some of the more tangible steps we have taken to operationalize our AI efforts. We've created a structure at GHC corporate that can provide resources to forward deploy into our businesses. These resources will accelerate projects, provide AI guidance to help organizations prioritize efforts and be a centralized resource for our companies to think through data policies, tool utilization and training programs. We are working with our units to: one, measure actions and processes by time spent; two, help prioritize building tools and apps that reduce that time spent; and three, working with our teams to understand if and how those hours should be reallocated.

One brief example. A monthly accounting close is required at each one of our businesses. One of our units have implemented AI tools as part of its monthly close process to improve efficiency. We've had great success, and we are in the process of leveraging this work across our other business units.

We also launched a Graham Holdings-wide AI Hackathon, where employees throughout the company can submit projects that they believe can improve business. The winning project will receive a $100,000 price. It has been encouraging how many of our units have leaned into this event.

The world of AI is evolving quickly, and it's imperative that we understand that and evolve faster than our competitors as we undertake these efforts, and we are operating with that mentality.

At this point, I'll turn it over to Andy, who will provide you with a bunch of updates on Kaplan.

Okay. Hello, everybody. Thank you, Tim. I have described in prior years how we try to build Kaplan, patiently one layer of earning power at a time in businesses and markets we understand with capabilities others find difficult to replicate. That approach has carried us through a test prep company, a higher education institution, a global educational services business and now into a period of genuinely rapid technological change.

The story this year is a continuation of the one I've been telling. With the pace picking up and the 4 main topics I want to talk about today, our performance, an immigration headwind, a portfolio decision and the work we're doing on AI are, in my view, all one story told from different angles.

2025 is a strong year. Revenue of $1.74 billion was up 3%. Adjusted operating income was $166 million, up 24%. On a reported basis, operating income grew 59%, but the prior year's figure was suppressed by a $22.9 million noncash impairment. By either measure, we produced meaningful operating leverage. U.S. Higher Education grew operating income 38%, driven primarily by Purdue Global PAUSE Supplemental Education grew 24% with gains across most of its program lines. Capital International grew 12% on flat revenues with strength in Australia, Singapore and U.K. professional more than offsetting the pressure in pathways that I'll come to shortly.

The first quarter of 2026 has carried that trajectory forward with each of our segments generating growth in revenue and operating income. In the quarterly comparison, the first quarter of this year included close to a $3 million boost due to the intricacies of how we recognize revenue under the Purdue Global agreement, and this will normalize in the second quarter. Setting that aside, the underlying performance is solidly ahead of last year across all of our segments.

A word about Purdue Global because the trajectory there is worth understanding. When we completed the sale of Kaplan University in 2018, the institution served roughly 28,500 students. Purdue Global now serves 41,000, the highest enrollment in PG's history, and growth of more than 40% from where that new university started. That growth reflects sustained investment by Purdue and Purdue Global's leadership in academic quality, student support, retention and careful financial discipline.

We've been glad to contribute our own strength, which, of course, have been meaningful to the story. But our choice of Purdue as a partner for that transition has aged well. I note some observers made snap judgments early that this relationship wouldn't work as intended. Fortunately, we don't pay much attention to the hot take crowd. The results seem to be a strength of the institution and its ever-increasing academic, social and financial value.

Let me turn to Pathways and Kaplan Business School or KBS, which are under real pressure right now and deserve a direct account. University Pathways is our largest single business within Kaplan International and a material contributor to Kaplan's earnings. New starts are significantly down year-to-date in both the U.K. and the U.S. and our deferred revenue suggests that we'll feel that pressure extend into the second half of the year. Kaplan Business School based in Australia has grown significantly in recent years due to its outstanding academic reputation and its appeal to international students, but it is facing similar headwinds.

When I spoke to you 2 years ago in this meeting, I described 3 structural forces, I believe, which shape our business for the next decade or more, a decline in U.S. high school students, the surge in global middle-class demand for higher education and the preference of transnational students for English-speaking destinations. I still see all 3 as major long-term drivers, but there's now a new policy layer sitting on top of all of them, at least for the time being. Governments across all 4 major English-speaking destination markets, the U.S., the U.K., Australia and Canada have moved in the same direction over the past 2 years, restricting international student flows. The motivations differ by country. The effect on pathways, Kaplan Business School and also our Ireland-based Dublin Business School is the same.

Now we've seen cycles in this business and in many of our businesses over a long period. We don't overreact to temporary shifts even once they may last a while. The pathway team has built and run this franchise for 2 decades with almost no down years through policy shifts, currency moves and a pandemic. The same people who delivered that record are running the business now. They are every bit as capable now as they were when they were creating that 20-year growth story that we've been enjoying. We don't mistake a cycle for capability.

We've also seen this particular pendulum before. Immigration policy in English-speaking countries oscillates. Restrictions tightened when migration becomes politically heated. They ease when the economic cost becomes visible. Universities lose revenue, local economies contract, companies struggle to find employees, tax receipts fall. Our business in Singapore, where the government has not pursued similar restrictions is growing nicely by every measure, student consensus, revenue, deferred revenue.

The underlying demand for quality educational -- education is intact, it is being redirected by policy, not extinguished. The geographic diversity we've built over years is doing exactly what diversification is supposed to do, carrying us when another region is under pressure. Pathways and to a lesser extent, KBS have carried other Kaplan units through their own tougher stretches in the past, today is the reverse. That is a purposeful part of how we've structured the company.

Our approach in the affected markets is straightforward. We manage costs carefully as volume soften, pulling out what is needed, but we protect the student experience and the partnerships we spent years building. Our students and our university partners should not feel the impact of a downturn that isn't theirs. We invest where we're not constrained. Singapore, our domestic higher education supplemental education businesses, online pathway options to carry less visa dependency. We expand source markets in the Middle East, North Africa, Latin America and Southeast Asia. And we position ourselves to recapture volume when the cycle turns.

As is almost always the case, we could manage this business for a better 2026 by taking actions that would compromise its long-term earning power. We're not going to do that. Long-term investors are not paying us to pop a single year at the expense of the franchise.

As Tim noted, we closed the sale of our Language Business last week. The reason he laid out in last year's annual report letter that are thesis for staying with the business through COVID and beyond hadn't been realized and that it was unlikely that the growth required to justify the necessary investment level would material -- would materialize is what led us to the sale.

I want to say clearly, this was not a decision about the Languages team or its programs. The Languages team delivered student outcomes in the very top tier of the industry and navigated the pandemic with a level of skill and integrity I continue to admire. This was a decision about capital allocation. Physical language schools run on fixed costs, leases, staffing, they require consistent volume to produce acceptable returns. The students paid materially less than students in our degree programs, which leaves far less cushion when volumes move against you. And the business is exposed to the same kinds of geopolitical and immigration shifts that I've been describing without the deep academic relationships and switching costs that make our pathways and degree business is more resilient.

The contrast with our other international recruitment businesses is worth pointing out because it speaks to how we think. Pathways, for example, is that this is under cyclical pressure with an intact long-term thesis. It's got a pre-proven team, 20 years of success and demand that is at least partially redirected rather than lost. Languages was a business where the thesis we've been waiting on simply didn't materialize. Same discipline in both cases, different facts led to different outcomes. I wish we had felt better about Languages prospects, but we gave it a very long look.

The portfolio remains what it was with one business now out of it. Kaplan's 2025 adjusted operating income, excluding languages, would have been approximately $183 million on revenue of roughly $1.6 billion. That's a solid base to build from, which brings me to AI and what we're actually doing with it. And let me first share my own thesis and it's just that a thesis that tells you one reason I'm optimistic about Kaplan's future.

Education is roughly a $7 trillion global industry, and the vast majority of that spending happens inside public institutions. The U.S. adds a substantial private nonprofit layer. The private for-profit sector where we operate is a relatively small share of the -- of that large total. It stays small because, among other things, education has strong protective structures, accretiation licensure, regulation, tradition, reputation, the scarcity of seats at selected schools. These structures exist for good reasons. They also mean money doesn't move quickly from one part of the system to the other. Students don't or can't switch education providers the way they switch streaming services.

AI is going to materially improve that education can deliver. Tutoring available at the moment a student needs it. Feedback that used to take days becoming continuous. Learning that adapts to an individual's pace and gaps. Products and experiences that were conceptually obvious for decades, but economically out of reach. These aren't marginal improvements. They change the value proposition.

Students will be major beneficiaries of AI, provided that the gains from better learning resources outweigh the temptation to use AI to skip to learning itself. That is a real tension and educators everywhere are working through it, but the net, I believe, will be strongly positive.

The economic gains will flow to whoever adopt AI most effectively to realize these improvements. Public institutions have real strengths, but their governance, their procurement systems, political considerations, faculty processes, not to mention cultural bureaucracy and caution was not built for rapid change. Private nonprofits have more flexibility, but still have plenty of similar constraints. Private for profits are more often built around innovation and nimbleness and they generally have fewer structural barriers. Not none, we still work with the creditors, regulators and partners, and we take those relationships very seriously, but fewer. And in a period of rapid change, that matters a lot.

So the thesis is this, over long periods, many years, not quarters, the gap between what the facets adopting operators deliver and what slower moving institutions provide is going to widen. The protective walls will hold for a long time, but the wall eventually yields where the value proposition gap becomes large enough and the share of global education spending flowing out of traditional institutions to demonstrably superior private sector offerings should grow. In other words, to be direct about it, I believe it's very likely that in the coming years, private companies will, as a group, outperform traditional education institutions as private companies tend to be faster moving and typically have less bureaucracy and fewer governance and cultural constraints. We're talking about a large pool of money and even modest shifts in how it flows are significant.

Now to be clear, none of this guarantees Kaplan anything. We still have to build products that are genuinely better, and we still have to compete for students, for partners for funding it more. We are not guaranteed a single dollar of revenue, but conditions will increasingly favor operators who could move quickly and we intend to be one of them.

I've spoken in prior years about our belief that generative AI will meaningfully reshape education. And then on balance, we see more opportunity than risk. A year or 2 deeper into the work, I can be a bit more specific.

Let me start with this. The cost of producing educational content is coming down, and it will come down further. But I don't believe it's headed to 0, at least not for content we put our name on. Quality educational material still requires careful human editing, subject matter betting, pedagogical judgment and ongoing oversight. What AI changes is what our people can produce with their time, not whether their time is needed. That principle runs through everything else I want to say.

Kaplan is a deeply human enterprise. Our core strength has never been about content or technology in isolation. It is our people to trust and respect they have for each other and for our students, and the fundamentally human work of helping someone get better at something hard. That is what our students and partners pay us for at the deepest level. That is a strategic asset, not a sentimental one, and we intend to protect it as we embrace AI. Our students assume that we will marshal the best technology on their behalf. We are not trying to become a one button company run by bots. And I don't think that would be a winning strategy if we tried.

A combination of AI and excellent people, our people empowered by AI not being replaced by it, is what will win. Our culture is one of our strongest assets and a key component of what we call the Kaplan Way is continuous transformation. We are accustomed to being nimble and embracing change. I believe that adaptiveness that adaptiveness will help us transition faster than most in the industry. The shape of our workforce will evolve, roles will change, and the mix of what our people do and the way we staff individual processes will shift meaningfully, which should not change is the character of the place.

We are on a path towards what we internally call AI First. By that, I made an organization where the default assumption about any processor activity is that AI handles it unless the activity generally requires human judgment, oversight or the uniquely human elements that our customers pay us for. Now there will be a lot of such circumstances, but our people will be vastly more productive and powered by AI than they were without it. We will not complete this transformation in a single year. And indeed, in some sense, I expect the process will go on without end. But we have made real progress over the past year, and we expect meaningful further progress by year-end and still more next year. The trajectory is one I'm pleased with.

The work falls into 3 categories, each creating value differently. The first is process works; reconciliations, content production, admissions workflows, adviser support. Here, AI is an automation tool and the efficiency gains are direct and measurable. We have staff assistant platforms live across all of our company advisory teams, for example, saving meaningful time for call and producing real cost savings this year as we grow student volumes without growing headcount. In many cases, we're using AI for time-intensive processes that are simply too time consuming for a human to do, but AI can fill the gap.

The second is cognitive work, advising, tutoring, analysis, content design. Here, AI is a capacity amplifier. The outlook still looks like a tutoring session, or a graded essay or a piece of advice, but it is produced faster, at higher quality and at a scale we couldn't previously reach. One AI tutor product is deployed across more than 70 courses serving more than 50,000 students.

KapAdvisor, our AI college advising tool has reached 100,000 cumulative enrollments and contributed to nearly a 17% growth in our pre-college segments. These are not pilots, they are scaled products changing how students engage with us.

The third category is one I'm particularly excited about and the hardest to quantify today. AI is letting us build products and services we simply could not have afforded to build before. Some that are new and some that had previously we had contemplated, but we're economically out of reach. We are working on several of these, and you'll be hearing more about them in coming years.

We have some structural advantages in doing this work well. We have proprietary data, decades of longitudinal student performance, behavior and outcomes that most competitors don't. Generic AI is widely available. AI combined with Kaplan's data produces results, others can replicate. And we have a diversified portfolio, which means that when 1 business unit builds something that works, we can deploy it across the others rather than rebuilding for scratch. The standard we hold ourselves to is not activity or experimentation for transformation, measured by what's actually changed in how work gets done. We are looking for scale deployment and measurable trajectory, and we are putting real capital behind it. The returns should show up over the coming years in both our cost structure and in new revenue.

Four topics. our results, integration challenges, portfolio adjustment and AI embrace one story. Across all of them, we're doing what we've always tried to do, build patiently, remain disciplined through cycles, reallocate capital when a thesis hasn't played out and move decisively on what we think we can see coming. The circumstances change, the approach shouldn't. There will be disruptions along the way. There always are. What we can control is the quality of our programs, the strength of our partnerships and the discipline with which we allocate capital, along with the character of the organization that does all of it. On each of these, I believe Kaplan is in good shape. Tim, back to you.

The meeting is now open to nominations for election of directors. We will then have the voting on directors followed by voting on the proposal before the Class A shareholders. The ballots are being counted, the floor will be open to any questions on business matters or comments you may have. It's now in order to proceed with the election of directors. There are 10 directors to be elected, 7 by the holders of Class A common stock and 3 by the holders of Class B common stock. The chair recognizes Mr. Cooney, who is a holder of Class B common stock and who is also a substitute proxy for a holder of Class A common stock.

I nominate the following persons for election as directors of the company to hold office until the next Annual Meeting of Stockholders and until their respective successors shall be elected and shall qualify or as otherwise stated, provided in the bylaws.

For election by the holders of Class A common stock; Tom Gayner, Don Graham, Jack Markell, Anne Mulcahy, Tim O'Shaughnessy, Rick Wagoner, Katharine Weymouth.

For election by the holders of Class B common stock, Tony Allen, Danielle Conley, Chris Davis.

I second the nominations.

There being no further nominations, I declare the nominations closed. I now declare the polls open for voting for the election of directors and for the proposal to approve the 2025 compensation awarded to named executive officers. They've been informed by the inspectors of votes with the holders of 96.7% of the Class A common stock have voted their shares. Therefore, the only ballots to be distributed will be those for the election of directors by holders of Class B common stock.

Many of you have already voted your shares for this meeting. If you have already voted online or by proxy card, your shares will be voted as you instructed, so please do not request a ballot now unless you wish to change your vote. Otherwise, tabulation of the vote will be unnecessarily complicated and delayed. For any stockholders who wish to vote by ballot, please raise your hand now, so the inspector of votes may locate you.

[Voting]

All right. If there were any, i direct inspectors of votes to distribute the ballots in after the voting, inform me whether they have completed the tabulation of the ballots. We will now pause briefly while the ballots are handed out. After they've been marked and been collected, we'll proceed with questions or comments.

While we're waiting for the results of the voting to be tabulated, we will open the floor to any questions or comments you may have relating to the nominees for election as directors or about the business operations of the company. I want everyone have a chance to ask a question or make a comment. So raise your hand, so I can call on you and tell us your name and who you're at, and we'll get going from there.

All right. We've got our Q&A period is open now. I see a hand there [indiscernible].

John [indiscernible] from Raymond James. I have a comment, throw a question and then possibly a real question. Nice to see the receivable from Purdue go down meaningfully, something I look at each year. The throwaway question is, have you ever considered changing the other category to another name. And the reason I ask is it reminds me of the big Aristotle, also known as Shaquille O'Neal -- who if you weren't a star on the team or he didn't know your name, he called you the others, okay? So I would -- I love Shaq, but I always thought that were condescending. So this is my throwaway question, obviously.

Have you ever thought about calling it future earnings or future value or something besides other?

Not really, to be totally honest.

Okay. Would you think about it?

If it's something that represents it better, maybe, but it's not. It hasn't been high on the list of [indiscernible].

I understand. And possibly the real question. You talked about in the annual report, and I think on Investor Day, about Graham Media. I don't think Graham Media gets enough credit for how great it's done over the last 5 or 6 years producing cash. The future looks a little tougher. But you talk about regulation and law changes and what does that look like? And I know it's a long shot, but we just kind of want to understand what we're hoping for to maybe have a better future in media?

Yes. So I'll start and then maybe I'll ask Catherine who's sitting over here to opine as well. We are very thankful for Graham Media Group. Catherine has done an excellent job running that business, or predecessors before that done excellent job running that business. And it has generated a lot of cash for the company that we've been able to use to grow Kaplan in to grow other businesses segments and it continues to generate a lot of cash for the company today. I sort of -- it's fortunate that Andy and Catherine are sitting next to each other because it's a little bit of like a boxing match of who's going to go and generate the most cash flow for the year this year. So we'll see how that plays out.

But the regulation question, there are structural challenges with the sector, and we've discussed those and I think many of those are known and obvious. And you have a set of laws that have been in place that were put in place decades ago in many cases, some of them going back to mid-century of the prior year in some cases. They don't make sense in today's date. If you may not like -- there's a lot of political narrative that gets tied up. But if you just take a step back and say, "Hey, is the way that this is structured, are the -- what does the competitive set look like? They really are vestiges of a different era. But there are still laws, and they're still part of how the industry has to go and operate with those regulations.

The -- my sense is that if you were to change some of those regulations, the ability to have more scale and buy more time to figure out what the industry looks like and what this local news and video product looks like in the future is enhanced. I don't think making regulation go away solves the problem. But I think it generate more time to figure out what things might look like on the other side. And it probably would involve with some of what you've seen. There's been incremental consolidation that's occurred, and there's a big attempt of that right now as well.

But -- so I think that's the output. I don't think it is an end all be all solution, but it is something that buys time to have a better shot at that, Catherine?

First of all, thank you so much for your comments. We are, like, Tim said, looking at the opportunities in front of us and the relaxation of some of the ownership rules give us a lot of optimism when we look at what the opportunities could be. Setting aside what's happening right now at Tegna and Nexstar, I think that everyone feels like this relaxation of the rules allows us to be able to look at the future from an optimistic standpoint, because it means that some of those rules that have been onerous, some of the things like Tim said, you're looking at a time gone by where we are competing against Southern media. And now we compete against everybody in big tech in a big way. And the rules just don't make sense anymore.

So being able to relax those rules and give us some opportunity to be able to get real synergy by partnering or being able to look at new acquisition opportunities or looking at our region and our markets as really unlimited potential in terms of what we can do for those communities is really something that we think is a strong possibility for us. And so we look forward to that. And I think the truth is, and I was just talking to someone this morning, our communities want and need what we produce. We just through AI looking at what Andy just shared with you, it's inspirational for us to say how can we apply AI to allow us to be able to better serve those communities with less friction with less production, being able to get those stories to those users and people in the communities that desperately need the kinds of information that we provide to them every single day. And so if I didn't have an audience, a community and customers, frankly, that desperately need what we do, it would be a different conversation.

Few questions under the broad bucket of capital allocation. One is going to the annual report from this year, you shared the operating manual, I think, for the first time, which is -- I think it's an aggregation of the way you guys have essentially operated. So one question there is one of the things you mentioned is to be flexible enough to issue shares if you believe you're getting more than you're giving. And I think looking at the recent history of the company, or I don't know if the company has ever issued shares to do a deal. So that seems to be something new and would love some thoughts around that.

And then on the more near-term basis, in the first quarter, you bought back around $35 million of shares after being dormant I think, for most of 2025. So perhaps that speaks to your thoughts around the growing earnings power of the company. So maybe you can talk a little bit about that.

And then I noticed you sold around $50 million from the equity portfolio, again, after a long period of being pretty dormant. So I don't know if the degree there to which you can discuss your thoughts around that. So few capital allocation questions.

Okay, so maybe I'll go in reverse order. We don't tend to comment on specific securities, but there was something that we thought made sense to sell at that point in time. And so we sold a bit. The share repurchase, it's the first time that we've repurchased meaningfully in middle, Q2, Q3 of 2024. So a long arc of time that a 16 to 20 months break is probably not that long. But we repurchased shares under a set of guidelines, which I've talked about before. And so those guidelines were all met, and that's where we began to repurchase. And as I mentioned earlier, we continued to purchase into Q2 as well. So we like that, and we hope to be able to do more.

On the comment on the potential issuance of shares, I think it's hard to have, but it's not a signal. If not, there's anything. But I think it's hard to be intellectually honest if you don't have that in your possible consideration set, we are trying to increase cash flow on a per share basis over extended periods of time. And if there is a scenario where you issued shares and you -- we thought we were getting more than we were giving that could be a rational decision to do so. So it was really just in that spirit and vein and making sure people know that we are not wedded to a share count that must consistently go down. We are wedded to growing cash flow on a per share basis.

And now I tend to think the circumstances are such where it is much more likely and that has been, in my 11 plus years of the company, it has been true that we have not found any of those opportunities, and we have found opportunities to bring the share count down. But it's really in that vein. I think it's hard to be intellectually honest with yourself and with everyone if you don't have that in your consideration set as well.

Question that came in just for the meeting, I thought it would address. So in 1970, our portfolio company Berkshire Hathaway created significant shareholder value through its participation in the friendly takeover of blue-chip stance. Given the precedent would the company be open to pursuing similar opportunities in the future, specifically engaging in strategic or activist positions within established publicly traded companies. I thought that was an interesting question. And in general, we owned a position in the company, and we had a good relationship with management and they wanted us to help. I think we would be open to that.

I think there are a lot of ways we can spend our time. And there are a lot of ways we can create value without a hostile and activist type of position in the company. So I think that is an unlikely scenario for us to have an unwelcome interaction with the company. A welcome interaction with the company, of course, we'd be open to that.

Mark Hughes with Lafeyette Investments. In your letter, you mentioned the purchase of what's now Hoover Architectural Solutions, you mentioned that you assumed $108 million of pension liability. Could you just review what an advantage that pension -- we've tried for years to figure out what to do with the overfunding pension plan. Could you just talk about what role your ability to assume those liabilities played in that transaction?

It was a key role in that transaction. The seller had an underfunded plan and that they were having to make annual cash contributions to. And this particular business, I don't want to speculate into their overall strategy. But this particular business, they had deemed as different [indiscernible] noncore from their operations. And so the fact that we could assume those liabilities was of a huge cash benefit to the seller. And for us, we wouldn't have -- we don't have to make any additional cash contributions. So a large portion of the value exchange was the assumption of those liabilities.

We have now done this twice in my tenure in company. In 2017, we acquired 2 television stations, WSLS and Roanoke and the CW and Jacksonville, where an assumption of liabilities was also part of that transaction. We would -- we're open and we would like to do more. So if anybody is sitting there with an underfunded plan and a decent business, give me a ring. As you might imagine, we've run all of the screens of public companies and have a decent understanding of what that looks like. We don't have as much insight into private companies and private companies also can have pensions, and they also could have underfunding issues. And so I think that's an area where something were to come about, we would be [indiscernible] kind of replicate this type of transaction.

Question for Andy. Andy, what explains the acceleration in enrollment growth at Purdue Global in the most recent quarter?

Excellent performance. I mean I think the reputation of the institution has continued to improve. I will say that the slowdown of the economy is helpful as well. There is no question that people, when they feel a little more concerned about their professional future, they try to -- they tend to invest in themselves with education. So that's probably part of it as well. But I think that Purdue Global is recognized -- the admired institution out there now among online institutions. And I think we're seeing the effects of that.

Yes, just to reiterate. I don't think there wasn't anything special. It was just a consistent kind of execution of the business.

Just a quick follow-on to that on the education side. So Andy, to do any of the university relationships beyond Purdue become meaningful? I think in the annual report, you talked about Wake Forest, UMass and I think a couple of others. But I know Purdue is the largest, but do some of those other relationships become consequential from a financial perspective over the next 3 to 5 years?

I hope so. I hope so, and we will continue to build them. And I think is a good chance that 1 or 2 or maybe more will break out over that period.

And then maybe transition to the Healthcare business, Tim. Thanks for the update. Actually, I was wondering about some of the operational results for the quarter, and thanks for sharing some of the infrastructure investments. Few questions. One is, any thoughts -- I don't know if our business leader for health care CR new business leader Dee, but would love to hear kind of her early thoughts on the business and her impressions and the opportunities, and I think you outlined some of those. So on the biologics side, I think that's a new disclosure that you just shared with us and is that a much larger opportunity over time than IVIG?

And I was also interested to hear that you mentioned that there's still opportunity to take share within IVIG, which is interesting. Maybe you can [indiscernible] that you feel comfortable scaling -- quantifying some of those opportunities over the next several years, it'd be really interesting to hear your thoughts.

And then just a bigger picture question is, CSI I still comping at 30% year after year, quarter after quarter. So when do the law of large numbers kind of prevent -- I mean, now you're introducing biologics, so maybe doesn't continue, but without giving guidance, maybe just share the magnitude of the opportunity over time because this just seems like a gold mine.

I'll start with the -- the infusion business, and then I'll turn it over to Dee, who is here, and so I'll turn it over to her to give some of her impressions on Graham Healthcare Group so far. So the -- on the CSI and on the biologics side, it is -- there are more disease states, it's a bigger market. It's a different market. There's different structures associated with it. And so we believe that it is a very logical adjacency. And we've, I would say, likely been exploring over time, but a very logical adjacency that leverages a lot of the same operations and processes that we have.

It was really a question of when do we want to meaningfully go into that. And we made that decision in, kind of the second half of last year that it lets go and pursue this business in earnest. And so there's a lot of disease states. There's a lot of growth associated with that, but there are some different structures that we need to make sure that we handle well.

On the broader growth opportunity, yes, we are continuing to invest to become a bigger business than we are today. And I specifically called out the fact that the sales team was 34% larger over a 6-month period of time and that they were kind of still in training or ramp phase because both have chosen the numbers, but to give you a sense of what we think the opportunity set still looks like. And what I would say is we started as a regional business that I would say is more super regional on the way to becoming national. So we are licensed everywhere, but I would not say we are scaled everywhere.

And the -- what is nice about that is we can look at our markets, and we have a sense, we're in enough places where we have a sense of how that -- those cohorts will grow. And as I referenced in my letter, I believe, then we are more markets where we are not close to fully penetrated than are. So we believe that there should be meaningful continued growth to come. The -- to take a step back, part of that is that this really makes sense. If you look at demographic trends, if you look at disease state trends, if you look at desire for in-home care to occur and infusion to occur and the compliance associated with receiving in home as opposed to having to go out somewhere, there's -- we're in a very good part of the -- of a pie that is growing that's part of a bigger pie that is also growing. And so when you have multiple tailwinds from the category you're in, if you can execute well, you should be able to be rewarded for that. And I think that's what we've seen and expect to continue to see.

I agree. You can only grow at 30% for so long, and we'll do our best to continue that growth in a responsible way. But we expect that we should be able to continue to grow at least in '26 and for the foreseeable future beyond. So Dee, do you want to give your?

[indiscernible] On the home health and the hospice business, which has now been the CEO for about 5 months, I've been incredibly impressed by the strength of this organization as I've come on board, particularly around the financial and operating discipline, the strength of the people and the kind of expertise and tenure of the team as well as the compliance culture that exists all the way down to the front line.

To me, though, there are the 3 ingredients of an incredibly strong foundation, and it's a great base from which to start my tenure. As I look forward, this business is poised for growth. We have a lot of growth actually left in the market and the service lines where we already operate today. There's density to be gained in the geographies where we have current footprint. Our hospice business particularly has a lot of potential for upside. And so the acquisition that Tim referenced is indicative of our view that there is still room to penetrate further where we are today. Beyond that, there is growth potential in geographies all over the country, frankly. And I think the hardest thing that we face as a team is what do we say no to? And how do we make smart decisions about where we plan our next flag pull from a geography perspective.

There's also service line expansion. I think that's a little bit of a further -- further play for us, but we continue to look at and are starting to -- kind of take our heads up and think about beyond home health and hospice, how do we take the strength of this organization and begin to add additional service lines that strengthen our proposition across the care continuum. So it's been a great onboarding, a great few months, and I look forward to what we can do with this organization.

[indiscernible] She's been great in her first 5 or 6 months at the company. We are really lucky to have her as the leader of the organization here. And I hope you just going to introduce yourself after if you get a chance to.

Tim, you said that conditions were met for a share repurchase in the first quarter of this year. What condition was not that last year?

Quite a good question, any way. I think the -- the opportunity set that we had and some things that we were looking at, we wanted to go and preserve cash for, and those things didn't play out and so the set of variables changed.

So just a quick question on CapEx guidance for the year. I think you're at $90 million to $100 million, which is a bit of a step up, I think, from previous years. I think in previous years, you were at $70 million, but that was offset by sales, I think. So I'm assuming a lot of this has to do with some of the investments you've talked about in health care, but are we talking about a higher CapEx level for the next few years as you make investments? Or am I kind of reading too much into the CapEx step-up?

Yes. We -- I think it's probably somewhere in between. The -- okay, historically, we've ended up maybe taking a little bit less than what our initial projections have been. I think if you look back over the last 5 or 6 years, that has tended to be the case. [indiscernible] have a couple of one-off projects that are in flight that are some reasonable size. There is a new kind of large. I think we're hoping it's a flagship type of restaurant within the group that should open up later this year at our Honda dealership as well. There is a large image refresh project that's going on there, too. Those are both one-off and not systemic, but most years, there is something.

There's always a 1 or 2 off type of items. So it's possible, it's up a little, and there might be a little more one-off-ish than normal, but I don't think it's directionally off.

The other piece I would say is that there is a little bit of CapEx in the -- historically in the Kaplan Languages Group business as it is a premise-based business, and that is also no longer will be part of the equation, which wouldn't have been in the initial forecast as well.

Tim, you said a few years ago that if we saw you spending more and more money on Framebridge, that was a good thing. You made a few comments about Framebridge. Can you tell us where you're at, what metrics you're looking at? Or -- so obviously, you're pleased with how it's done because we're spending more money. So just a little bit more detail on Framebridge, what you're seeing and what you like and I know you're not going to answer this, but when will it turn positive? So that's my question.

So we love the business. We love our place in the business, and we love the market that we occupy. It is a bigger category in space than most people realize on the surface, and we have created a differentiated experience, both from an online and a retail standpoint that I think has a lot of more characteristics to it and it would be very, very challenging to compete with over time.

The -- with the other thing, we also have a formula at this point on retail locations where we can be pretty smart and understand where places should go, have a relatively good idea of what that store should be able to produce, and we are starting to accelerate our ability to go and roll those out. We are early in how big this business has a chance of ultimately getting. We are closer to the finish line than the beginning of the capital consumption phase associated with it. We're still in that, but I think we are beyond peak, and we are at a spot where operating leverage with future revenue growth really should start to happen.

The 2 metrics that we really look at are: one, the demand generation side, which will be mostly driven by store growth as well as same-store results. I don't -- that is a very clear picture to me. We have 47 stores right now. We're not in your state, Florida at all. There are a lot of markets that we need to enter there. I wouldn't say there's a single metro area where we consider ourselves penetrated in terms of store rollout as well. So there is a long way to go on that front, but a very clear line of sight of how we get there. And I just point that out because it's so rare in business where you have a level of confidence in the demand side and that, that will continue to happen. And I think that both I have that and Susan has that confidence.

The other key metric is how much do we keep for every unit that we produce or the gross margin. And that is the area where we're really focused on. And that is the key toggle when we're sitting here 10 years from now, have we been -- has this been an okay use of capital, a great use of capital or knock it out of the park use of capital. And so the level of driving the efficiency with which we produce at our studios is the other kind of key metrics. So that store growth and then how efficiently we produce in those gross margins are the things that we really are focused at driving internally. And something Susan and I spent a lot of time talking about both of those things.

So I think the opportunity set is intact. We are -- clearly see a path, I would say, on both of those things. I think one of them is very -- I want to say, rent and repeat that doesn't give enough credit to it and the other requires some additional step changes in how we actually produce at this point in time.

So it appears that it's a first mover advantage that there's not any unreal technology into framing that I'm aware of. And I've asked you this question before, but to get into Florida, to get into more states, and we're doing it all by ourselves. It seems like that slows us down a little bit. Any thought of capital partners? And I know that's maybe last for me to you. But to speed up the process so that we're 120 locations in 2 or 3 years. What's the up and down of that?

Well, I would -- I like your 120, because I would -- if we're sitting here 3 years from now, I would guess we'd probably be about 120 locations. There's a world where we should have a rhythm where we are opening a cadence of 25-plus stores a year. I would hope you are at that rhythm in 2026. So -- and certainly, that rhythm in the back half of 2026.

Capital isn't a constraint on it. The -- and one of the other things that has been asked on occasion is their franchise model that exists. And that would be -- we're very challenged on that because this is -- these are having a safety culture and the chain of control associated with that is hugely important when you're dealing with things that are valuable to people both could be from a financial standpoint or it could be the movie tickets from the first date of their spouse that they went on with their spouse that died. Like that is valuable in a different way. And having the right systems and care and controls over that is something where, in a franchise world, that always has been something we -- a bridge we wouldn't be comfortable with.

And so that is a core part of how we think about differentiation is and what makes it very, very hard for others to compete with. And Kaplan is [indiscernible] at this point in time either.

Other questions. We've got time for just -- we have 10 more minutes if there are questions. I think there's Rajiv's hands going up there.

I wanted to make sure there was no one else before I kept on going here. But Tim, on the manufacturing division. Maybe talk to -- it seems like business is getting better. If you exclude Arconic, which obviously affects how you look at the numbers, but can you just talk us through the improvement that we're seeing in the cash flows from that division. I know some of this, you've talked about in previous years have been various macro issues affecting each of the 4 subdivisions. But maybe talk us through what's going on broadly with that segment?

I think Hoover is the biggest within that segment, what I would call the majority in the traditional Hoover business on the fire retardant wood side is tied to multifamily housing, among other things, that is still in a down cycle. The team is running that well and is figuring out how to produce a good amount of cash in that world. And I'm really excited to see what it looks like when we get back to an up cycle there.

The -- but I think we're -- it appears that we're through the worst of that, that some of those trends, at least in the last couple of quarters have not continued to get worse. The -- on the rest of the business, we're really -- we completed a plant expansion at [ Joyce ]. So that was one of those one-off CapEx projects from a year or 2 ago. That has worked well. We are seeing -- there was a consolidation of 2 facilities plus an expansion that allowed us to take advantage of some incremental demand and some customer dynamics in the industry, and we took advantage of those. And we're reaping the benefits of that decision right now.

The last one I would comment on is that Dekko, their biggest end market was commercial real estate. That was a terrible, terrible a few years. For a while, it was when would it stop getting worse. We reduced the cost structure on that to -- if we were to return to previous levels of revenue, we would have a better business. We are starting to see some growth there. We've gotten that commercial cycle. We've gotten through that, and you're starting to see some growth in that end market again. And I would say that's the last thing I would comment on that we're trying to see some of the benefits flow through with a little bit of incremental revenue there.

Other than Framebridge, could you maybe talk about the businesses in the other segments that are still losing money and the general plans for bringing those to profits?

I would talk about -- and how we always view this. So with an investment stage business, do we think the capital is going to be something that will go and drive the acceptable returns off of that investment that's going in. And so that's how we analyze. And what we do is if something isn't working, we -- or if it could be a better home or somebody can do a better job with that, we will look at that. So last year, an example of that was with World's of Good Brands, where the amount of referral traffic from search engines was really impacting that Digital Media business quite substantially, and we didn't see an ability to turn that around, and we came to the conclusion that needed to be -- those brands should be a part of something that has more scale. And so we decided to exit and sell that.

And so that is an example of we couldn't see it half, we will take that action. And that's how we view everything within that as we monitor, engage with those businesses pretty closely. And it's -- I think we look at it how I think you hope that we would look at it where something not a permanent place there if we don't see a path where the returns on the capital invested will be adequate.

My last question, I promise. This one is on the pension plan. I just wanted to clarify. So I think it was -- it went from roughly $2.5 billion over funded to $2.7 billion overfunded roughly at the end of 2025. So that's even after the Arconic outflow. Is that correct? I don't know.

Sounds correctional, I guess.

And I know there were some thoughts around various things you were thinking about in previous years, including taking money out and it didn't make sense at the end of the day. And then you talked about looking for other opportunities. But any other thoughts I know you tried experimenting with the Healthcare division, perhaps using some of the overfunded pension to retained workers, whatever program was thought -- didn't quite work out. But any other kind of ideas currently, I think you guys are thinking about from a pension perspective and [indiscernible] that you have?

Yes. We are always thinking about are there ways to leverage the pension for -- of employees and as a strategic asset of the company. And I would say our -- we figured out new ways to do it that we hadn't thought of 5 years ago or 10 years ago, and I would suspect 5 or 10 years from now, there will be new things that we haven't thought of yet. This is not a well-tried path. The -- we are -- when we look at something, there's not case law, there is -- there's not a lot that exists where you can say, okay, we should do X, Y and Z. So in many cases, we are doing what I would call the original work and analysis and talking with various different experts, whether it's governmental, legal, et cetera, to understand what is and isn't allowable, getting opinions and perspectives from them. And that work is what has allowed us to incrementally figure out how to leverage it over time, and that work continues.

So I think that is -- it's an important and a huge benefit to the company. You're looking at a lot of active pensioners in the room right now, and we like that. But we know that we will, over time, figure out how to leverage it further, so.

Okay. All right. Well, the polls have now been closed and the ballots for the election of directors and the proposal for the Class A shareholders have all been tabulated.

Mr. O'Shaughnessy, the inspectors of votes have presented their report showing the following results: on the election of directors to hold office until the next Annual Meeting of Stockholders and until their respective successors have been elected and shall qualify or as otherwise provided in the bylaws, the following directors have been elected by the holders of Class A common stock and all received 928,001 votes. Tom Gayner, Don Graham, Jack Markell, Anne Mulcahy, Tim O'Shaughnessy, Rick Wagoner, Katharine Weymouth. The following directors have been elected by the holders of Class B common stock, all received at least 1,630,311 votes. Tony Allen, Danielle Conley and Chris Davis.

On the proposal to approve the 2025 compensation awarded to named executive officers, the holders of 928,001 shares of Class A stock or 96.27% of the outstanding Class A stock voted for the proposal.

I hereby declare that based on the report of the inspectors of votes, the individuals nominated by the Board of Directors and named in the report have been duly elected directors of the company. In addition, the proposal to approve 2025 compensation awarded to named executive officers have been approved. I direct that the report of the inspectors of votes be filed with records of the meeting.

Is there any further business to be brought before the meeting? If not, I suggest we adjourn.

I move that this meeting be adjourned.

All those in favor of the motion, opposed, motion carried. I hereby declare the meeting adjourned.

Good afternoon, and welcome to the Graham Holdings 2025 Investor Day. Please note the presentations at this Investor Day may contain forward-looking statements that are based on the company's current expectations. All public statements by the company and its representatives that are not statements of historical facts are forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995.

The forward-looking statements are based on expectations, forecasts and assumptions by the company's management and involve a number of risks, uncertainties and other factors that could cause actual results to differ materially from those stated, including, but not limited to, the company's business strategies and objectives, the prospects for growth in our various business operations, future financial performance, and the risks and uncertainties described in Item 1A of the company's annual report on Form 10-K and quarterly report on Form 10-Q for the period ended September 30, 2025. Any forward-looking statement made at this Investor Day speaks only as of the date on which it is made. The company assumes no obligation to update any forward-looking statement after the date on which such statement is made, even if new information subsequently becomes available unless required by law.

A full disclaimer is posted on the Graham Holdings Company's website, along with the meeting presentation.

Please let me introduce Timothy J. O'Shaughnessy, CEO of Graham Holdings.

Hello, everyone, and welcome to the 2025 Graham Holdings Company Virtual Investor Day. We are delighted you've chosen to spend an hour or two with us to learn more about the business.

I'm Tim O'Shaughnessy, the CEO of Graham Holdings, and I will be your master of ceremonies for the day. My colleague is Andy Rosen, the CEO of Kaplan; and Catherine Badalamente, the CEO of Graham Media Group, will be on the line as well and available for questions.

I will walk through a slide presentation with corresponding remarks and then we'll open it up for a Q&A session, which will last until about 2:30 Eastern Time, or until there are no more additional questions.

2025 has been a very year for the company. Our biggest businesses have performed at or above our expectations and our largest investment stage business has started to gather steam.

We continue to manage the business with a focus on driving cash flow on a per share basis over the long term, inclusive of non-consolidated investments. We do this while always maintaining a balance sheet that lets us sleep well at night, knowing the unforeseeable is always a possibility.

We think about driving free cash flow per share in a few primary ways: One, owner's earnings growth. This reflects an increase in the underlying earnings power of our businesses; two, share repurchases. We care deeply about the price we pay. We will not repurchase shares under a blind philosophy to shrink the share count. However, when our shares trade at a meaningful discount to our view of intrinsic value, we can purchase shares from time to time to increase our free cash flow per share in an efficient manner.

Three, pension. Replacing expenses funded from our corporate treasury with pension eligible expenses. An example of this is replacing a corporate match of a 401(k) plan with a pension benefit for employees. While this reduces GAAP net income, our cash flow grows by the amount of 401(k) match, our corporate treasury no longer funds.

And fourth, acquisitions. Over time, we can add to free cash flow per share by acquiring businesses. This occurs in two ways. We may occasionally add a new platform business, but most of our acquisition capital is likely to go to bolt-on acquisitions that help us grow our existing segments.

Many of you likely know something about the company already; some may be new. In either case, you'll be pleased to note that the above formula when combined with a focus on providing reliable products and services to our customers, has stood the test of time and will continue to be our playbook moving forward.

Let's talk about a few key updates from the year. In February, we spent a little over $200 million to redeem the majority of the mandatorily redeemable noncontrolling interest, or MRNCI. In layman's terms, this means we spent quite a bit of money buying shares from a group of minority shareholders of CSI, our specialty pharmacy home infusion business.

This past July, we acquired the since renamed Hoover Architectural Solutions business as carve-out from Arconic. This is a leading aluminum cladding manufacturer for North American non-residential architectural and retail markets.

In October, we acquired a Honda dealership directly adjacent to two dealerships we already owned in Woodbridge, Virginia.

Last month, we completed a refinancing of our $400 million notes that were due in June of 2026, along with our revolving credit facility.

And finally, a few key personnel updates: Spiro Roiniotis joined our corporate staff this summer as Chief Technology Officer. Spiro has worked at different units of the company since 2009. I asked him to consider joining our corporate office as CTO to help our organizations think through their technology strategies and specifically, to help some of our businesses accelerate their leveraging of AI to drive improved results in their operations.

Last week, Dee Grein started as the new Graham Healthcare Group CEO, leading our home health and hospice operations. We are thrilled to have Dee join the business. Most recently at Optum, Dee had a long tenure helping to grow health care services operations and shaping the strategy that allows for long-term success. We cannot be more thrilled with her appointment.

Results year-to-date have shown how the business has evolved to be less impacted by the cyclicality of election cycles at Graham Media Group. Through Q3, our adjusted operating cash flow has increased by $3 million, from $307 million to $310 million. This is despite a $45 million year-to-date reduction at Graham Media Group. These improved results have been driven by Kaplan, Healthcare and Manufacturing. While we have managed to stay ahead of last year through 3 quarters, we are unlikely to end the year ahead in adjusted operating cash flow. The election spending in 2024 peaking in October and early November make the comps a hill just a bit too big to climb.

Kaplan has had an excellent year, navigating visa complexity at Kaplan International to minimize impacts on both students and financial results. and continuing to strengthen our product offerings at KNA. Revenue has increased 4% year-to-date. Adjusted operating cash flow has increased 18%.

I'd like to speak for a moment about the Kaplan Supplemental business. You can see from the financials that revenue is up 9% through Q3 and adjusted operating cash flow is up 21% over the same time period.

This business is an excellent example of how a model that includes patience, focused on serving customers, and a long-term time horizon can lead to an excellent result. It wasn't that long ago that the majority of revenue in this segment came from physical, in-person Kaplan Test Prep centers. This was a leaky bucket that largely went away by the end of the 2010s. The team has spent much of the last 5 years, repurposing our skill sets and capabilities to become the online leader in test preparation and professional certifications. While the team was brilliant at carefully managing costs, unfortunately, the revenue was leaking faster than new revenue was coming into the bucket.

That will end this year. For the first time in many years, this segment has organically grown its top line. The team at Kaplan has spent the better part of a decade creating a unified management structure, a centralized technology system and a centralized marketing function to execute upon a sources, extensions, and connection strategy that allows us to provide services to a student throughout many phases of their academic and learning journey. Every dollar of revenue is hard fought; every new product is not guaranteed to be successful. It's very, very hard to turn around and reshape a business at scale, while maintaining a reasonable level of profitability throughout the turnaround. Greg Marino, Steve Marietti, Lee Weiss and the rest of the team have really done something remarkable.

Graham Media Group continues to manage the changing landscape. In 2025, the anticipated decline in cash flow due largely to reduced political spend has materialized. In addition, other challenges remain.

In preparation for this event, I reviewed last year's remarks and noted we discussed several negative trends challenging the business, primarily around the erosion of the linear ecosystem and the fragmentation of audiences. With the passage of another year, those trends have not improved. We also spoke about the possibility of deregulation providing a new opportunity to compete. In 2025, it appears regulators may be more open to exploring those conversations, but only time will tell.

The local broadcast industry is quickly approached point of no return, where the diminished audience and economics could make many markets non-viable under the current regulatory framework. Society and regulators should ask: would we rather have one or two news operations in a market that have a fighting chance to continue to serve their local communities for the foreseeable future? Or would we rather have four that may be destined to fail? The current top 4 rule and the traditional definitions of market share makes 0 sets in today's world.

If that's not enough, the industry's network relationships have evolved to one where stations pay more for less. Out is the spirit of partnership that define network affiliations for decades. In is an approach for networks to maximize leverage in such a way that many stations have no choice but to cut news operations in order to pay network affiliation fees. To add insult to injury, networks use those fees to help pay for content that they make unavailable to their local station partners. Local stations have a legal obligation to serve the public good. The current dynamic makes it increasingly challenging to do so. Within a few years, all but the largest markets may no longer have an economically viable model, and those markets could ultimately be on the same path as well.

No matter how this plays out over the next few years, Graham Holdings will be okay. We've grown our business in a way that ensures we have a strong balance sheet and cash flow coming in from a variety of operations. In most scenarios, our total company earning power will increase over the coming years. But we used own The Washington Post, we've seen the erosion that can quickly take place in a local news industry. Foolish cross-ownership rules that stayed in place far too long accelerated the decline of the newspaper industry not too long ago.

The inflection point for the industry is upon us. Absent substantial regulatory changes occurring in short order, with timely subsequent transaction approvals, the industry will enter into a managed decline. In this scenario, Graham Media Group will continue to generate cash in the short to medium term, but the long-term prospects are challenged.

Our Healthcare segment has continued its admirable growth rate year-to-date, with revenue increasing 36% and adjusted operating cash flow increasing 46%. JV income has been flat.

As mentioned previously, we're thrilled Dee Grein has joined lead Graham Healthcare Group, our home health and hospice business, comprised of both wholly owned units and joint venture operations. She takes over a business that has continued to perform well with both census growth, strong patient outcomes and good financial results. While the rate environment remains challenged, the team is doing an excellent job of driving indirect cost efficiency to maintain good results.

James Sheets and the CSI team have continued the excellent patient care, service to doctors, and strong financial results. Specialty pharmacy is hard. Providing care in the home is hard. Doing both of these things well is challenging and requires great operations, both clinically and pharmacologically.

We are still relatively early in our CSI journey. Many of our more recent areas of geographical expansion are still underpenetrated relative to some of our more mature markets. We recently entered California in the last few months, which represents 12% of the U.S. population, although our operations are very early stage. Our share of the IVIG home infusion market remains relatively small, but we believe we can continue to gain share of a growing market over the coming years. We'll continue to invest through the P&L to achieve this future opportunity. In 2026, we will expand our team to handle higher patient census. Additionally, we will build out several new pharmacies that will allow us to increase future patient volume. With those elevated levels of investment, we still expect we will be able to meaningfully grow profitability in 2026.

Our manufacturing operations have improved with adjusted operating cash flow increasing year-to-date from $30 million to $40 million. Capital expenditures have also come down meaningfully as our facility expansion at Joyce was completed. Our general trends have been improving with the exception of Hoover, which continues to be in a cyclical downturn due to the persistent down cycle in multifamily housing construction.

The automotive operations continue to perform well. Adjusted operating cash flow decreased from $34 million to $28 million through Q3. As expected, we are below peak earnings achieved in the early years of the decade and corresponding supply chain disruptions. The results are also down this year due to weak operating results at the Jeep dealership that was closed in September 2025, transaction costs associated with the purchase of a Honda dealership and incremental investment in Roda, our valet service operations.

Our other businesses continue to be a mix of profitable, unprofitable and investment-stage businesses. As always, the mix within this segment evolves over time. Year-to-date, the segment has been roughly flat on both revenue and adjusted operating cash flow. However, what's going on under the hood is more encouraging and puts the segment on the cusp of revenue growth and improved adjusted operating cash flow. Let me provide a bit more insight.

Earlier this year, we divested the World of Good Brands media operations and wound down the remaining infrastructure. By Q4, the remaining costs associated with this were modest. World of Good had seen declining revenue and meaningful losses that will no longer burden the P&L.

Society6 is in the midst of the last stage of its cost structure evolution, which, when completed, will result in a business where the downside has been minimized. We expect 2026 results will show sizable improvements.

Clyde's Restaurant Group has a large presence in downtown D.C. and extended government shutdowns have historically temporarily depressed sales. This shutdown was no different. While, I'm sure this isn't the last shutdown we will experience, we do not think it is standard.

Framebridge continues to be the largest investment within the segment. The news is largely good. Here are some details: the core metrics at Framebridge are mostly headed in the right direction. Revenue was up both online and in retail due to new locations and positive comps at existing stores; gross margin has expanded; customer NPS remains near or at all-time highs; we increased our store count by about 40% in 2025; overall operating margin is on track to improve by several thousand basis points from 2024 to 2025.

As a reminder, Framebridge, as we have been building it is a business that requires scale. There are three main areas where investments have been required to support the scale we envision with the business.

The first is the corporate function. This covers SG&A costs such as technology, marketing, merchandising, finance and other related areas. While these costs will continue to grow somewhat from here, they are now scaling at rates substantially below the growth rates of both revenue and gross profit.

The second is our footprint of production studios were the framing occurs. As we grow our retail footprint, we need to expand our studio footprint to support the increased volume. For example, we were unable to practically support retail expansion into California or the Southwest United States until we had a production studio in place. We recently opened a studio in Nevada that supports this expansion. While we anticipate opening additional production studios over the coming years, including an additional facility in 2026, these costs are also now scaling at rates substantially below our revenue and gross profit growth rates.

The third is our retail footprint. We have the opportunity to open many, many more stores than the 44 we expect to have open by year-end. These stores require capital to build out as well as investment to support preopening expenses and training. In 2025, we will have opened 13 new stores, short of our target of 20 to 25. We are disappointed that we did not achieve our targeted store expansion. Delays in store openings increase our total investment required to get to positive cash flow. However, the 4-wall model for our stores continues to show very strong results. And as we get more experience, we are incrementally improving results with new openings. We are once again targeting 20 to 25 store openings in 2026, and our retail expansion costs are now scaling at rates substantially below our revenue and gross profit growth rates.

Let me spend a bit more time about why I'm optimistic about the long-term economic profile of Framebridge. First, as a reminder, we believe we provide a superior customer experience. We can win on quality, price, speed and convenience. The customer value proposition is compelling and repeat rates and NPS scores validate that belief. When a customer uses Framebridge, we tend to become their exclusive solution and they also begin to frame things they had not previously thought to frame, growing the overall market opportunity.

Second, we are investing capital for expansion by opening up new regional pods. A regional pod is composed of a production studio and a set of retail stores that will leverage that studio. There is an upfront capital expenditure investment required to build the studio in those stores. Our investment case is based on a belief we can achieve attractive cash-on-cash returns on the upfront capital investment once the pod has been fully built out.

Here is a map of where Framebridge has stores today. If you notice what I noticed, there's a lot of empty space. What do Miami, Denver, Seattle, Phoenix, Tampa, Detroit and Orlando have in common? They are all top 15 DMAs that currently have 0 Framebridge retail locations. In fact, New York and the D.C. area are the only metro areas with more than four locations in the market.

If we could snap our fingers and have 10 more regional pods tomorrow, we would. Unfortunately, not quite that easy. On the retail front, location, location, location remains as critical as ever. Finding the right stores at acceptable lease terms remains essential. With the production studio, we need to find the right space with the right labor pool acceptably close to the retail stores it will service. And we need to do those two things in concert with each other, so under-absorption at the production studios does not persist for an unacceptable duration.

We've not been investing or building Framebridge to be a small business. If we thought the realistic annual revenue for Framebridge at maturity would be in the low or even mid hundreds of millions, we would have taken a different approach. As we enter 2026, we expect Framebridge to continue to drive operating leverage in its business, and the overall investment level should decrease.

Now that we've covered operations, let's spend a bit of time on the balance sheet. As of September 30, we had $1.236 billion in cash and securities against $732 million in debt, leaving us with a net cash and security position of $504 million. In short, we feel very comfortable with our current liquidity.

We also recently did something we don't do very often, access the credit markets. In November, we closed on a new $500 million bond offering and a $400 million revolving credit facility, the proceeds of which were used to pay off $400 million in notes due in June of 2026, and as well as pay off our term loan A obligation. The new notes mature in December of 2033, and we are pleased the credit markets were so receptive to our offering. I'd like to thank Wally Cooney, Mac Greisler, Elaine Wolff and the rest of the team for their efforts. The specific primary terms are listed in the presentation, and I will let you peruse at your will.

Earlier, I referenced the acquisition of what we now call Hoover Architectural Solutions. We love this type of transaction. We acquired a complementary business to one of our existing operations, and the majority of the purchase price was funded by the assumption of approximately $107 million in pension liabilities. Those pensioners whose liabilities came with the transaction are now in a much better spot than they were previously.

Post transaction, our pension plan funding ratio was not materially changed, but the cash flow profile and earnings power of Hoover certainly was. To put a finer point on it. You can see our estimated funding ratio and estimated absolute funding levels as of September 30, post the transaction.

With that, this will conclude the prepared remarks portion of the event, and we'll now open it up for the question-and-answer session. [Operator Instructions]

First question is, I'm curious about the recent addition of Cable ONE shares as revealed in the 13F. Value of the company is a fraction of what it once was. Has management considered acquiring Cable ONE in full?

So first -- and we got a couple of questions on this. So that came in advance. No, we have no interest in acquiring all of Cable ONE. In general, we aren't going to comment extensively on securities because of the history here, I'll maybe give a bit more than I normally would. Business we followed for a very long time. And so know well, know the people well.

And at a very high level for a business that we feel like we know well, I think both maximum balance sheet concern and maximum competitive threat pressure have both been priced in by the market, and we thought in a way that wasn't particularly rational. So it is a straight investment calculus, and there's not much more to it than that.

The next question is, during 2025, the company acquired two businesses, one in Education and in Manufacturing for $19.9 million in cash and the assumption of $107.4 million in net pension obligations. The assets and liabilities of the companies were acquired -- of the companies acquired were recorded at their estimated fair values at the date of the acquisition. In the case of the Arconic, a significant portion of the purchase price was funded by the company's assumption of certain pension obligations. How financially meaningful is this new business in the context of GHC and the Manufacturing division? And this is the largest such transaction using GHC's overfunded pension. I know you're always looking for such opportunities. Are there more such companies out there?

So I guess on the first portion of that question, the business is not -- it's meaningful to Hoover. It's important to the Manufacturing line, and it doesn't reach a level of materiality to Graham Holdings on the whole. So maybe that's a little bit of a helpful context for people.

On the second side, are there such -- are there other opportunities out there? I hope so. We would love to do more transactions like this. It is -- it can be a bit challenging because you need somebody that has an underfunded pension and views that as an issue and the seriousness that they should. There needs to be a business that we want to own over an extended period of time as part of a transaction, and we have to know about it.

And the third is often the hardest. If it's a company that's a public company, the -- we can run screens, but there are many private companies that have had pensions, and there are lots of private companies out there, and that's much more challenging to understand. And actually, the carve-out transaction that we did here, Arconic was a private business. So it came to us through an adviser channel. And so that is something that the private world is harder to ascertain what's available out there. So if people know things, have suggestions for us, we're always open. We would love to do more transactions like this.

All right. The next one is the Education division. Why has adjusted cash flow improved so significantly year-to-date for supplemental education and higher education?

Maybe I'll give a quick answer and then turn it over to Andy here. Some of what I mentioned in my remarks is the reason. It's a lot of -- the bringing together to create KNA really created an operating structure where we've been able to kind of accelerate product development and improve margins. And then the higher ed side, the Purdue relationship, our ability to book the full fee has started to occur more recently.

But Andy, you might have more to add on that.

Yes. Thanks, Tim. That's actually an excellent summary of the reasons. There's nothing very unusual, but rather an exam prep, as you say, it's a product of intentionality and discipline, I would say. KNA Supplemental business is a whole portfolio of businesses and products, which each have their own market dynamics and cycles. But we pulled together a shared service function that is able to deliver real excellence on some of the core functions, product development, sales, marketing and so on at a lower cost per unit than the individual businesses, and that's helping on the expense side and it's helping on the revenue side. And that combination is a good one.

And we're in markets where we think we can grow for some time to come. So that's a good story. And on Higher Ed, as you say, Tim, Purdue Global has been growing and improving, and we've -- it's profitable enough to fund our full fee and in fact, some repayment of accrued fees. So we currently -- over the course of the first 3 quarters, we averaged 37,000 students and which is about 5% more than the same period last year, and we're currently standing at about 38,500. So it's the basics, growth and expense discipline that's leading to more cash, nothing more dramatic than that.

Okay. Well, thank you, Andy.

The next question is, I was a bit surprised by the amount of Graham Media Group deleveraging in Q3, revenues down 28%, but operating income down 57%. Anything new to know about or just normal non-election year dynamics?

It is primarily non-election year dynamics. You really see a lot of in election year spending in Q3. That is -- those are revenue dollars that are highly accretive. And so when that disappears, you get real deleveraging. There is -- the other trends that I mentioned in the remarks continue to exist, but the primary reason for it is we had a very nice election cycle, and that did not repeat in 2025.

Okay. Next question is CSI. So in August 2025, CSI purchased Pine Drug Holdings and was issued a California pharmacy license with dispensing operations expected to commence later in the fourth quarter. You foreshadowed CSI entering California at the annual meeting. Is this a huge 2026 boost to CSI? Another question today, 70% of CSI is IG. Is that a good proxy for the go forward? And then how is CSI still comping so nicely?

So some of this was covered a little bit in the remarks, but yes, we are now operating in California. It is -- we are in the early, early stages. So I would expect that '26 will have some business. It is -- calling it a huge boost would be an overstatement there. It don't instant get to scale. It takes time to build up teams and operations. So we're excited to be kind of fully geographically penetrated, but it will take a while.

On the IG is 70% of CSI, is that a good proxy for go-forward question. Look, we expect to continue to grow the IG area at a nice clip and in line with what we have shown in the past. I do think that there's the opportunity to incrementally go into other areas. So I think we can grow IG, but I think that there is the set -- the opportunity to expand a bit in terms of treatment lines. And if we were to do that, it naturally would go down a little bit as a percent of the overall.

And I think -- how is it growing so nicely? We are continuing to expand geographically and penetrate further into the markets that we've been at. And we are really good at getting -- providing service for our doctors that we work with and getting patients on census and getting them treatments in really efficient periods of time and making everybody's life easier as part of that process. So I think we've created an operational process that's working really well and I think has allowed us to continue to get referrals in this space because we really make our doctor, partners' lives easier.

Okay. Moving next. The year-to-date through third quarter cash flow -- operating cash flow is up despite GMG being such a drag. It now appears that GC cash flows can grow possibly even in a non-election year and is the $80 million to $90 million CapEx guide a good proxy for '26?

Look, at a high level, and I addressed this a bit, but the -- our growth, excluding Graham Media Group, we expect to continue. And in election years, we expect Graham Media Group growth to show over the prior year. So the business that is less cyclical, I think we've had reasonable growth, and we expect reasonable growth to continue for the foreseeable future. And at some point, as those other parts of the business become bigger, they may offset the down years in non-election with Graham Media Group.

And as far as the CapEx guide, we'll put out a good proxy for '26. I don't think there was anything extraordinary about '25. We had a couple of big projects that wound up. So -- but you never know what new projects come up along the way as well. So I don't think it was abnormal, but there were a few projects that wound up.

Okay. Overall thoughts on immigration policy changes in terms of the impact on Kaplan International. KI growth appears to have slowed on the top line. Andy, I will kick that one over to you.

Yes. And you're putting your finger on the exact reason that growth has slowed, which is we have immigration policies in a number of our student receiving countries that have become more negative towards immigration. And countries are rarely opposed to student immigration. They oppose immigration more broadly, but student immigration is the lever that's easiest to pull when they feel political pressure. So we're seeing that in -- not just in the U.S. and the U.K., but also and even more dramatically in Australia and Canada. We've been at this for a long time, and these are cyclical. You can't really imagine a world in which Canada and Australia remain anti-immigration for a long period of time. These are huge countries with small populations. They need people.

But politics plays itself out over time. And I think the immigration policy issues are going to persist over the course of some period of time into the future. The Chinese economy actually has resulted also in fewer students to U.K. pathways and some of our other businesses. But you think about -- in the U.S., for example, presumably because of the lack of a trade agreement with India and the continuing negotiations, there are essentially no visa -- U.S. visa appointments available in India, which means that all the students who want to come to the United States, including a lot that we've recruited, can't come. And so that -- those are just people who are willing to pay us and our U.S. university partners money that are stuck in India because we can't get them into the country.

Australia, as I said, is -- continues to tighten its regulations. But I would say KI has been -- Kaplan International has been really effective at finding solutions to problems and diversifying our student base. And we do have some advantages. For example, Singapore is a net beneficiary of all of this because students who can't get into Australia or the U.S. or the U.K. are going to Singapore now, and we're picking that up. So it's a mixed bag. I expect we're in a -- we're definitely in a relatively tough period, but that's going to turn around over time.

Okay. Thank you, Andy. Next question. Nice job on the refinancing and upping the credit line. Can you remind us on thinking of how much financial leverage you're willing to bear?

I think first, it's just important to say, I think we're very comfortable with where our balance sheet is. Our net cash and securities position is referenced a little over $500 million as of the end of Q3. And so we have a lot of optionality. We are never going to be a company that really thinks about leverage as something that is an accelerant in a way that could add risk to our overall enterprise. That is just not how we operate. And if you're a shareholder and an investor in the business, you should really understand that.

The -- I think the first thing that we do is when we're looking at capital is, what are -- where are the possibilities? We generate cash from operations, and we use those. But we also have -- we have a balance of securities that we could use to go and fund any acquisitions or large outflows of cash. We have a revolver that we can use for -- if there's a shorter-term reason to go and do something.

So if we went through all of those and we also -- or if it didn't make sense to leverage those for some reason, we would -- for the right deal, we would add incrementally to our current leverage profile, but I don't think we would add transformatively. And we don't really have a target that we operate against. But we're going to remain in a spot where we're very comfortable and sleep well each and every night.

All right. Given the rising demand for early childhood education, particularly in foundational literacy math and bilingual learning, does Kaplan see early-age learning as a strategic expansion area? If so, what capabilities or partnerships is Kaplan considering to enter or scale in this segment? Andy, can I kick it over to you again?

Yes. I mean to the extent that you -- by early child education, you really mean preschool. I don't see Kaplan -- that's not a current target market for us. There's so many other issues you're dealing with beyond the educational -- the core educational issues. And we're not particularly expert at those issues and maybe the time will come when we feel like we're ready for that, but that's not yet.

Having said that, I think we do think that the K-12 arena might be an appealing area for extension of our offerings. You may have heard we acquired a small private K-12 school in Florida this year called Ohana. And we're investing in that school, and we think it's an opportunity to learn from and contribute to a thriving K-12 school community. And so that gets us in the neighborhood of what you're talking about. But I think that early childhood education is probably not -- is not at the top of our list.

I couldn't have said it better myself. All right, the next question. Do you hope -- do you see any hope that Congress will pass a bill to allow you and others to use excess pension funds for non-pension uses? Or do you think the 50% penalty for taking out pension funds will be with us for many years to come?

Well, I think the question had the phrase, hope that Congress in it, which might provide an initial view of my answer. Look, we're not going to have the ability in any real way to help drive what Congress does. So in no way do we view a legislative solution to a pension overfunding as something that is in our realistic sphere of things that we think about.

So our primary usage has been how can we leverage the pension funds to offer benefits to employees, improve the risk profile of pensioners. Sometimes we're bringing new pensioners into the fund from underfunded plans via transactions. And figure out are there additional pension eligible expenses, which allow our treasury to have less. And so that -- less expense.

So that is going to continue to be our playbook. It is -- there is -- as the questioner references, there is a 50% excise tax on any withdrawals and then you also pay local and federal taxes. So there is quite amount of tax leakage associated with an overfunding withdrawal, which is why we have not taken that overfunding out and have instead looked for ways to operationalize it. And until something tells us that there is a different path on that front legislatively, I think that will continue to be the path that we move forward on.

Okay. Next question is, how strongly have you considered selling your TV station business given the long-term secular concerns you have? You don't sound too hopeful about the business long term. So why not sell it and use those proceeds to buy a different business?

I think the -- I'm not sure our view on the TV business and local broadcast, as I articulated here, is all that different than what the market's view would be. If you look at multiples in the space, they have come down over time, and they don't project -- the multiples in the space don't imply a big growth opportunity. So I'm not sure we're that misaligned with the market on the whole.

Look, if regulatory change happens, and I think I've said this before, we'll look -- if the landscape change, we'll relook at the landscape. And so that remains true today. We do have -- when you think about a transaction, there are things that we think about in terms of we have a pretty low tax basis on that business. We own the properties outright. And I'm not sure the market gives you credit for those things.

And so -- and last and most importantly, we have a good management team that's running it well and generating a lot of cash today. So -- and the discount value associated with that cash in the next year or 2 is going to be a pretty low discount rate. So as the world shifts, we'll continue to look and watch. We love our balance sheet, which gives us an opportunity to participate in any way that would make sense. But the landscape in that world is changing in a space that is getting harder and harder. And so of course, we'll continue to be -- as possible regulatory change happens, we'll continue to evaluate if there's a different way in which we should participate in the space.

Okay. Next question. Would you consider opening a Framebridge studio in Northern Idaho, Montana or Spokane, where there's infrastructure and trades talent.

That is a question that I would not have the specific answer to. The Framebridge team would have a clear view on where they're evaluating. But I can tell you, it needs to be within a reasonable distance to enough major markets where the retail stores can be serviced well enough. So that's a key factor along with labor pool availability.

Okay. Can you speak to the exceptional growth at GHC, which businesses are stronger than others? Are there M&A opportunities? Or do you see the business continuing to focus on organic growth? Does buying in minorities change how the business allocates capital? Okay. There's a lot in there.

Let me piece it apart into two different buckets, kind of the businesses and then capital allocation. Our capital allocation rubric has not fundamentally changed. We look at where is the dollar going to be spent that we think in a perfect world improves our moats and generates the best return on capital. And so that's really how we view things. The -- we can do this in a variety of ways, organic investments, acquisitions, share repurchases, dividends. And many years, we do multiple -- most years, we do multiple of those things. Most years, we don't necessarily do all those things, but most years, we do multiples of those things. And so I expect that will continue.

On the M&A front, I think it is more likely that our capital will go into what we call bolt-on acquisitions internally. That tends to be more likely than it does that than kind of a brand-new business in a brand-new category, although we don't rule that out as a possibility. But the reason is when we do a bolt-on, it is in a business that we are in or complementary to already. So we tend to know quite a bit more. The management team that is going to run it are people that we know and have worked with. And oftentimes, there are some cost structure benefits to the new company that's better by being owned by us.

And so you put those things together and the kind of the risk-adjusted returns on capital usually look pretty good relative to a brand-new business at the parent level. So I expect over time, more of our M&A-related capital that goes out would be on that bolt-on size, but not exclusively, but just more.

On the buying in minorities, the other part of kind of that is also a capital allocation decision. And I view it as a slightly different form of share repurchases. We are actually buying more for shareholders of the businesses that we own already. In some cases, it might just be one business. And so while we did not have much on the share repurchase front through Q3 in 2025, we did put a lot of capital out buying minorities and increasing the share of earnings for shareholders of those minority businesses. And we may do that from time to time. There are not that many places where we don't own 100%, but there are some, and there may be opportunities that make sense at some point in time.

And then the last piece of it, the growth at the company overall and our -- which businesses are stronger than others. We find this a very challenging question to answer because different businesses have different characteristics to them. Look, I think that our scaled operations largely tend to have earning power coming from multiple different sources. And that, in my mind, is usually a good thing, not a bad thing because if you have -- if a bad thing happens at one of Kaplan's product lines or business units, it doesn't mean that Kaplan is going to have a terrible year.

And if something were to -- if we had at Graham Healthcare, our Healthcare segment, the home health care rates for 2026 are actually modestly down from 2025. And this is in an inflationary environment. But that's one part of the business, and it actually means that broadly, we expect to continue to be able to perform and grow through that. So I think I would characterize it as our larger businesses that have cash coming from multiple different business lines tend to be things where you're going to see a level of kind of strength and durability. But we have some very, very good businesses that are not the top of the scale as well. They just don't punch as high in terms of their weight.

Okay. Tim, please speak more on who is running the CSI business, why were they chosen and their experience with CSI or health care in general.

So I referenced in the remarks, there's a gentleman by the name of James Sheets, who is the CEO at CSI Pharmacy. He is the founder of CSI Pharmacy. So he has been -- he is a pharmacist by training, and he actually has been running the business since its founding and has continued to run the business after we acquired it about 6 years ago. So he is running it. He is continuing to run it. And I look forward to continuing to work with him to continue to grow the business over the coming years. So he knows that business better than anybody else at Graham Holdings, and he probably knows the space better than anybody else at Graham Holdings.

Okay. Can you speak further about what exactly is driving your strong ongoing health care revenues with or without acquisitions? Are the acquisitions you may be looking at in this area growing as fast as your organic revenue growth?

So the health care growth this year has been largely, if not exclusively organic. We have continued to grow the business by really focusing on a high-level thesis of in-home care and both believing that, that is a good trend societally for patients, for payers, for hospital systems, and it is a -- has favorable macro demographic trends. And if you can build a business that you want to run effectively for 10, 20, 30 years or more in that space, you can take a long-term view. You can have excellent quality, you can have excellent operations and you can build towards that.

And then ultimately, you can hopefully take share. And so that has been the mentality that we've had. And we've been on this journey in the health care services space for about 13 years or so at this point. And so I think you're just seeing the result of a lot of work by a lot of people and participating in the right sectors of the space. We expect that all of those macro trends I referenced before should continue to be true. And so hopefully, we'll be able to continue to grow on that front.

From time to time, there will likely be opportunities to add inorganically as well. We've done that from time to time in the health care space. We will likely continue to do that from time to time in the health care space. And so I don't think it's an either/or though. We should be able to grow organically, and we should be able to find inorganic opportunities from time to time.

All Right. For Catherine, will the bump from political advertising be muted for 2026 given the dynamics in your business? Or will you still expect to see large election year political spending?

So the good news for broadcasters is that we're actually expecting record-breaking midterm spending next year. We're looking at it actually going to be the most expensive midterm on record and nearly half of it going to broadcasters. The one thing to keep in mind is that the spend is actually more concentrated to battleground states. And the good news for Graham Media Group is that we happen to be in one of those battleground states with Detroit, Michigan on record last time we ended up being with Detroit, the second highest political spending at our station in Detroit in the last election cycle. And so we expect, again, that to happen again next year. So between Detroit and what we're seeing in Houston and San Antonio with the Texas races, we expect a really great election year and a lot of political spending in our markets.

All right. We are excited about Michigan. So let's -- from Catherine's lips to the financial statements. Okay. Next question, how does Kaplan protect itself or take advantage of AI? Andy, why don't you take this one?

Sure. Well, that's a pretty tight summary of one of the core questions that we face. One of the things for sure that in education, like really anywhere in the economy, the world is going to look a lot different in 3 years, 5 years, 10 years than it does right now. AI is going to change a lot about education. And so the job of any of the players in that space is to move as quickly as possible to where we think it will be in the coming years. And we think we have a lot of -- we think we have a head start in that on a lot of fronts. And so that's a good thing for us.

But we can also be sure that there will be players we've ever heard of approaching things in ways we've never thought of who are trying to head to that spot as well or move that spot. And so we've got to be very conscious of that and keep a sharp eye out for competitors.

But adapting Kaplan for a world of AI is a central part of our priorities. And we really view AI's impact across the spectrum. So it -- firm on the one side, turbocharging productivity for individual employees or teams or functions. But on the other end of the spectrum, really enhancing and innovating on our products and services and building a whole new product opportunities for students. And we've been rolling out solutions across the spectrum, including things like AI-powered staff assistance and academic content creation and admissions conversion on the sort of left side of that spectrum. And then tutoring and program support and grading and real-time learner feedback and all kinds of things to help student engagement and outcomes on the other side of that spectrum.

So we're moving quickly on AI. And we feel like we're excited about the opportunities, but it's a fair -- it's absolutely essential for us to constantly be surveying the landscape to see who's doing things that we just never thought of that we've got to protect ourselves against. So yes, how do we do it by focusing on it every day.

And I would just like to say that Andy and his team have really done an excellent job of having an AI-first culture at their organization. And it's something that many of the other businesses at Graham Holdings are looking at them and admiring how they have been able to really incorporate it into their day in, day out operations to drive business value. So one of the examples and benefits of having multiple businesses is sometimes you see one of our operations does things really well, and the rest of our operations can start to learn from that. And Andy and his team have been leading the way at the company on that front.

The last question in the queue. So if others have questions, you should put them in now. But what does a managed decline scenario look like for GMG? So Graham Media Group is -- Catherine and her team are doing a really good job of running that business. And I would expect that we would generate meaningful amounts of cash over the short and medium term. If the dynamics of the business persist and don't change, that ability to maintain and protect profitability will get more and more challenged because the trends of the business on fragmentation of audience, negotiated contracts are worse than they were 5 years ago and are -- don't give any signs of getting better at the moment.

So I think what we will need to look to do, absent other changes in operations is look to protect profitability as we continue to manage the business. And that's what we do in all of our businesses if they have down cycles. If we have a down cycle that we don't see when there's an end to, we'll just continue to manage that as best we can and look for new opportunities to get into new business segments that could drive more durability. But we have a premium on making sure we're doing the right thing for our shareholders while managing the business and producing the best product that we can along the way. And so I think a managed decline scenario is when we don't think the business is going to grow, we'll do what we can to protect profitability while still putting out a product that we could feel really proud of.

Okay. How does the company view its marketable securities holdings against paying down long-term debt? For example, are there market prices where it might prompt you to realize the taxable gains and pay back some or all of the debt?

Yes. I mean, I think sort of a couple of questions embedded in there. The recent bond offering we did was at price at 5.625%. And so I would expect our return profile and marketables over any extended period of time, I think we would be pretty disappointed if we didn't beat that as a number. So if you just view it on a straight sort of rate basis.

The second, as you referenced, we've got -- we have large embedded gains within many of our securities. And so you would need to any -- whether it would be used for a debt paydown or something else, any usage of that, we think we would -- that usage would need to take into account the tax leakage associated with selling a security that had a large embedded gain. So -- and that, I think from time to time, we'll find that. And if we do, then the securities are things that we would sell, and we've sold some from time to time as well.

So it really is a decision around what is the return profile that we think we would get from the dollars that we would get as net proceeds. And how does that compare to our view of those securities and the tax leakage that you would get via selling them. And then a usage of that, whether it's debt or whether it's something else, is sort of a question tied to that return profile.

Okay. Next question. What does it take or what level of revenues or earnings does it require to move a company out of the other? Is Framebridge or Clyde's expected to move in the future?

It's an interesting question. For those that have been around for a long time, you may realize that health care was once in there, automotive was once in there, and manufacturing was once in there. And so that's when I referenced that, that category changes. It's not -- it's dynamic. It is a constant evolving mix of businesses over any extended period of time. Usually, something will move out of there when from a reporting requirement basis, it needs to become its own segment. I would say that is the high-level governing philosophy that we would have on when something breaks out and becomes its own segment overall. There can possibly be other reasons that you would do that, but that would be the primary driving reason.

Okay. Most information you've given about the Framebridge opportunity, how big is the market?

Well, the custom framing market in the U.S., there have been some third-party publications from a few years ago that referenced $7 billion to $8 billion in annual sales in the U.S. I would expect because of inflation and plus some of the things that we're doing to help start growing the market that, that number is a bit bigger than that today. So we -- it is -- I think it would be not unreasonable to assume that today, it's in the neighborhood of a $10 billion a year annual market. Our belief is that there is no brand in the space and that if you look at a lot of other markets, if you're a brand leader, you can end up owning some double-digit percentage of that market. So I think at the highest of levels, those are what our aspirations are with Framebridge.

Okay. Was a tax loss garnered on the World of Goods shutdown? If so, any chance to use against sales of portfolio holdings?

Not really because they were operating -- if there were operating losses, that reduces our taxable income in year. So it reduces at that point in time. I don't want to exclusively say that there wasn't some modest amount, but nothing substantive associated with that. And the other thing I would note is we were able to sell some of the brands and receive some proceeds associated with that in both 2024 and in the first part of 2025. So there was some cash that came in as part of the shutdown as well.

Okay. Is there any concern with Graham's exposure to Berkshire Hathaway given Buffett's retirement?

Not really. It's a -- Berkshire remains a pretty good business with a really strong balance sheet and cash coming in from a lot of places and excellent management that's been there for a long time, and I suspect continue to -- will continue to be so. So we don't really have any concerns -- any more concerns about exposure today than we did a year ago or 5 years ago and will probably a year from now.

Okay. There are no more questions in the queue. So I would like to, at this point in time, wrap up our presentation and thank everybody for attending. Once again, we appreciate you attending the 2025 Graham Holdings Company Virtual Investor Day. Thank you.