At the beginning of the presentation posted on our website, and now, I'd like to introduce Graham Holdings Company Chief Executive Officer Timothy J. O'Shaughnessy. Tim, the floor is yours.
Hello, and welcome to our 2024 Graham Holdings Company Investor Day. Our agenda will be as follows: a general update on operations and share repurchases, Wallace Cooney, our CFO, will provide an update on our pension plan, and then a Q&A session for as long as time permits. We've received several questions in advance and would like to remind you that questions can also be submitted real-time through the presentation portal. I'd like to start just with a first principles review of how we run Graham Holdings. Our high-level philosophy remains largely unchanged. Our goal and North Star is to grow our cash flow on a per-share basis over the long term, including through our holdings of non-consolidated investments. We also place a high premium on returning capital to shareholders when it makes sense relative to other available opportunities.
This has happened in the form of a quarterly dividend, which has and is likely to continue to increase over the years, and via share repurchases, something that has happened more often in recent years. I'd like to discuss our share repurchase program. Graham Holdings does not have a set repurchase cadence, nor will management likely ever propose that we do so. We only buy back shares when, one, doing so does not put our balance sheet at risk or severely limit our balance sheet flexibility. Two, internal investment opportunities that help grow and strengthen the moats of our businesses and grow long-term free cash flow have been funded. And three, when we can acquire shares at a material discount to our view of intrinsic value. Over the last several years, all of those conditions have been met.
There are no guarantees they will continue to be met in the future, and there will almost certainly be extended periods where this will not be the case. We'll continue to evaluate share repurchases based on the above-listed parameters. Since 2020, share repurchases have been a significant use of capital at the company. We have spent $580 million on repurchases and acquired more than one million shares through September 30, 2024. A shareholder who owned part of Graham Holdings on January 1, 2020, would have seen their ownership stake in the company grow by approximately 25% without ponying up any additional cash to achieve that increase. With the recent appreciation in the share price of Graham Holdings, the return profile of those repurchases looks increasingly attractive.
Based on the stock price at the end of November, the returns range from a 14% annualized return to a 37% annualized return, depending on the year of comparison, excluding the impact of dividends and excise taxes. Most would consider these good short-term to mid-term results, but we know in the long term the scorecard of the repurchases will be determined much more by the next 15 years rather than the previous five. Operationally, year-to-date results have been more than adequate at most units. Through Q3, we've generated $307 million in adjusted operating cash flow, up $52 million from the prior year. The improved results have been driven by Kaplan, Graham Healthcare Group, and political advertising at Graham Media Group. Results were partially offset by declines at our manufacturing segment, which saw several businesses in down-cycle markets.
The Q3 balance sheet snapshot shows a cash and securities balance of $1.1 billion against $765 million in debt for a net position of approximately $343 million. We continue to believe that we have sufficient balance sheet flexibility. These numbers have all previously been disclosed in either our earnings releases or a 10-Q, so I'd like to spend more time discussing our current views on strategy at several of our major businesses instead of doing a more detailed walkthrough of consolidated numbers. I'll start with local broadcast by highlighting that the strong political season we expected has materialized, with adjusted operating cash flow up 28% for the first nine months of the year. Political revenues in Q4 further exceeded our expectations. Our approach at Graham Media Group has not materially changed. We are taking the cash flow generated from the stations and reallocating it into other opportunities.
While we are doing this, Catherine Badalamente, as discussed at last year's Investor Day, is attempting to rebuild the plane in the air while maintaining acceptable levels of altitude. The task to maintain this altitude is formidable and not without turbulence. Here is what we believe to currently be true about the industry. Challenges: The linear ecosystem continues to erode. MVPD and satellite subscribers have been declining at mid to high single-digit rates year over year, causing net retransmission revenue to decrease and audiences to erode. Virtual MVPD providers are a partial offset to the T
inear ecosystem erosion, but with inferior economics and lower viewership. The average age of a local news viewer continues to trend up. Increased advertising options exist for national advertisers, particularly via Connected TV and programmatic channels, making direct buys less likely.
The product has not evolved in line with society and in most major markets is undifferentiated from the competition. The strengths: Local broadcast has community and convening power that continues to be very powerful. The ability to reach local audiences at scale still exists. Likely voters remain over-indexed to local TV, providing continued opportunities for political advertising. The brand awareness of stations and communities is enormous. Local advertisers continue to have success in driving brand reach and sales on linear, streaming, and digital. Competition in other local mediums, predominantly newspaper and radio, continues to weaken. Catherine and the team at GMG are leveraging our strengths and assets to build and test new models attempting to find the next chapter of a sustainable local media business model.
One item that would help: The FCC currently has broadcast ownership rules that prohibit a single owner from owning more than one of the top four affiliates in a local market. The origins of this rule date back to 1941, while the current version of the rule, in substance, was adopted in 1999. I imagine most attendees today are accessing and listening to today's event via some form of internet connection. In 1999, as a reminder, internet household penetration was approximately 26%. It's fair to say that times have changed in the last 25 years in how people access and consume media and content. The elimination of the Top-Four Rule would allow for local broadcasters to create healthier operations via mergers, leaving bigger operations that could better serve the community.
The additional scale would also provide more resources to compete against national competitors who do not use spectrum to deliver content. This would insulate local news-gathering resources and provide more runway to explore future models. But it is by no means a panacea. The future has not yet been invented for the industry. To ensure this additional runway is put to good use, the FCC could consider capitalization rules that would limit the amount of leverage that a spectrum licensee could take on if they own more than one of the top four stations. This would ensure that capital is available to truly innovate for the future. Removal of the Top-F our Rule does not solve the industry problems, but it would go a long way in helping to provide the opportunity to continue to serve the communities in which local broadcast operates.
Kaplan is a great example of a business that has successfully rebuilt itself on the fly. Andy Rosen and his team have constantly evolved Kaplan's business model in line with changing modalities, regulations, and student needs. Academic outcomes have always been at the heart of the Kaplan culture. Because of that, over the course of a generation, Kaplan has been able to adapt from a physical center-based test prep business to a scaled global business, with an increasing focus as a service provider to higher education institutions worldwide. Results year-to-date have shown continued progress in building the incremental earning power in Kaplan's businesses that we strive to achieve each year. At the annual meeting, Andy spoke a bit about the trends that drive the Kaplan strategy, with a particular focus on three specific areas. I'll repeat them here. First, the number of U.S. high school graduates will decline.
It's easy to predict the number of high school seniors we'll have in 10 years. Assuming immigration or mortality don't change dramatically, you have a pretty good basis for calculation in the number of second graders we have now. Since birth cohorts have been declining for many years, we can be fairly confident that today's U.S. kindergartners will graduate high school alongside about 3.5 million others, down from 3.9 million today. In addition, over the last seven years, the percentage of U.S. high school graduates who immediately went on to attend college has dropped from 70% to 61%. For parents of young children who are already thinking about where their kids will go to college, this is good news, as there will be less competition for space.
For a test prep company, the reduction in both numbers and competition is not great news, but it's at least as bad for many universities that will have difficulty filling their classrooms and dormitories. We believe that will create opportunities for organizations that can help fill those classrooms with, say, international students, or promote specific universities to high school students, or enable alternative revenue and delivery options like online programs. Second, international higher education students will surge. From 2020 through 2030, it is expected that more than 1 billion people will be added to the global middle class, mostly in Asia and Africa. Global tertiary education enrollment will grow by about 160 million students over the same period. Yet, there are too few local, high-quality universities to meet the demand in countries around the world, which will drive some of those students to seek their higher education in other countries.
Third, English-speaking countries are the first choice for transnational students. Approximately 50% of transnational students are captured by the Big Four English-speaking countries, which include the U.S., the U.K., Australia, and Canada. An estimated 80% of transnational education tuition dollars are captured by the Big Four. Again, Asia is a driving source of transnational students, with China and India making up a large portion. We will continue to build our global recruitment network and product suite to meet these needs. This will largely occur via organic initiatives, but it is possible we will, from time to time, find a business that would help us build out our moats faster and increase earning power. Lastly, Kaplan continues to invest heavily in leveraging AI and LLM models to improve its product offerings, as well as the efficiency of our operations. There are certainly both opportunities and threats that come from AI.
However, the decades of human experience, combined with the data from billions of student learning interactions at Kaplan, should provide us with a leg up as we leverage these new tools. 2024 has been a spectacular year at Graham Healthcare Group. Adjusted operating cash flow has increased year-to-date from $34 million to $54 million. Our share of joint venture income generated another $10 million year-to-date. This growth was led by improvements at our home health operations and CSI Pharmacy, our home infusion business. We believe our healthcare operations should continue to grow organically in future years and can become much bigger than what you see in 2024. It is important to note that we do not currently own 100% of CSI Pharmacy. We originally purchased the majority of the equity of the business.
Separately, we have a mandatorily redeemable non-controlling interest, or MRNCI, related to the ownership portion of a group of minority shareholders based upon an agreed set of terms. Changes in the MRNCI are recorded as interest expense and subsequently reflected as a liability on the balance sheet. This is a real liability, and you should treat it as one. This arrangement exists in several businesses, although the lion's share of the balance is tied to CSI Pharmacy. Through the first nine months of the year, we have recorded expense of $85.1 million to adjust the fair value of the MRNCI, which has increased our total balance to $125.5 million. The vast majority of this balance is scheduled to be redeemed in March of 2026.
In other words, we are currently scheduled to buy out an additional portion of the minority interest of CSI we do not currently own in about 16 months, and it is most of the current $125.5 million balance. We disclosed the details of this arrangement in several filings, the most recent of which was in Q3. The MRNCI balance will likely change between now and March 2026. In fact, there's a good chance that if CSI continues to grow and perform well, it will increase. The more valuable CSI becomes, the more it will cost us to buy out this MRNCI liability. In fact, if this were to happen, it would hurt our reported net income, and it would reduce our reported book value. But this would be a good thing. Why?
Because it means the much larger majority ownership stake we already do own would also have increased substantially in value. Due to accounting rules, our mandatorily redeemable minority stake needs to be adjusted in value regularly, whereas our majority stake lives on our balance sheet at cost, and its value cannot be adjusted upwards. We don't make the accounting rules, but we do, from time to time, find ourselves in a spot where explanations of oddities are required. Most of our service lines focus on providing in-home care, and we expect that to be our primary focus in the future as well. Our strategy starts with continuously looking to increase the quality of our care and outcomes. As we are successful with this, we should be able to further penetrate our existing service lines organically while opportunistically looking at bolt-on M&A opportunities.
Additionally, we expect we may add new service lines that can leverage our operating capabilities and/or geographic footprint, primarily via acquisition. The services we provide are a win for society. In-home care often provides better care outcomes in a lower cost of care setting than an acute care facility. Patient satisfaction is almost always higher when they receive care at home. We have an operating engine that can handle more cylinders. We're actively searching for new opportunities to deploy capital within our healthcare services segment in the coming years. My update on our manufacturing and automotive segments will be brief. Our approach is simple: manage the businesses well, organically try to grow our market share, and seek out the occasional bolt-on acquisition, particularly at our automotive dealerships. These segments produce cash that can often be redeployed elsewhere within the company or returned to shareholders.
While we have a few businesses in our manufacturing segment experiencing the downside of cyclicality, we are delighted with our managers and their ability to continue to produce solid returns on behalf of shareholders. Now, let's take a moment to discuss our other businesses. In aggregate, I expected more progress from this group in 2024 than materialized. At most of the units, results have been in line with expectations, but not all. I'll first start with what's going well. Clyde's Restaurant Group. Our restaurant group continues to provide great experiences for customers and create brand experiences that meet the current consumer moment. Additionally, we recently opened two new locations in the second half of the year: the Rye Street Tavern on Baltimore Harbor and Cordelia Fish Bar in Washington, D.C.'s Union Market neighborhood, which opened in mid-November. Both restaurants are through their pre-opening and into their ramp-up periods.
We expect they will be meaningful contributors to the group. Framebridge. We are seeing strong demand signals for our offerings at Framebridge. Our retail rollout has accelerated, and we expect to close 2024 with 32 open stores, up from 22 at the beginning of the year. Additionally, our recent launch of tabletop frames has added a new product line to our offerings. We are seeing strong signs of customer interest in our tabletop collection, particularly as a gift-and-go option. A customer can walk into any Framebridge store and AirDrop a digital photo to an associate's iPad. The associate will digitally crop and touch up the photo on the spot, then print and frame the photo in the store before sending you on your way with a now-framed photo nicely packaged in a gift box, all in just a few minutes. You never need to show up empty-handed again.
As we head into 2025, Framebridge will continue to use its growing scale to improve its unit-level margins, as well as open another 20-25 stores. Saatchi Art. Our original art marketplace continues to grow well. When the housing market begins to unthaw, we are optimistic we will see further increased demand as refreshing and/or filling walls with art tends to be correlated with a change in residents. What's not going well? The biggest surprise of the year was a large decrease in search traffic to many of our digital media brands after Google's March search update. Notably, this update began to surface a generative AI widget above many of the search results. I'm certain many of you have noticed this in your own searches. This update, combined with the new AI results, led to a serious traffic decline at several brands and units.
World of Good Brands was most significantly, but not exclusively, impacted. In a multivariate world, it's hard to calculate the precise impact of these search changes, but we believe this development led to a $10-$20 million increase in operating losses for 2024 relative to our expectations heading into the year. The decrease was so dramatic for a few brands that further declines would not be impactful because the base is so much smaller. Brand advertising spend has been below overall ad industry projections. Ad spending was forecasted to grow at higher rates in 2024 than had materialized. Correspondingly, our media and agency businesses have had more tempered results. Society6 continues to see headwinds in the home decor market. Significant reductions in the cost base have helped, but we need to return to revenue growth soon to see a sustainable business.
As discussed previously, this group always evolves, and we expect it to continue to do so. 2025 will continue to be an investment year for other businesses, although less so due to possible portfolio management, reduced pre-opening costs at Clyde's Restaurant Group, and improved results at many units. We will continue to monitor loss-making businesses with a view of risk-adjusted returns on capital. We've had some activity within our pension fund, most notably an annuity transaction that took place in October of this year. I've asked Wally Cooney, our CFO, to update you on how we think about managing the fund in ways that we believe benefit current pensioners, retirees, and the company. I'll turn it over to you now, Wally.
Thanks, Tim. Today, I'll provide an update on the current status of the company's pension plan, elaborate on recent activities, and share some thoughts on opportunities for the future.
As a refresher from previous discussions on this topic, it's important to highlight certain fundamentals with respect to the company's pension plan. These are included in the slide. The company's pension plan and its overfunded surplus is of significant long-term value to the company, its employees, and retirees. It has funded meaningful pension benefits for thousands of employees in the past and present, with the expectation that robust pension benefits will continue to be funded in the future that provide the company with a strategic advantage in recruiting and retaining key talent across our business lines. The plan has also funded thousands of employees with generous severance and early retirement pension-funded benefits, enabling our business units to make cost reductions as needed without impacting the corporate treasury.
The plan's substantially overfunded status leaves us with significant optionality and flexibility in exploring, designing, and implementing programs, plans, and possible transactions to effectively utilize the plan for the benefit of current and future employees and the company. Over the last few years, we have continued to look for ways to further de-risk the pension plan and to utilize the plan to provide current and future employees and retirees with meaningful and attractive employee benefits that are also beneficial to the company. Let me start with a high-level overview of the company's pension funding status in recent years. Since the end of 2020, the pension plan funding ratio went from 2.6 times overfunded to 3.4 times overfunded at the end of our last fiscal year. At December 31, 2023, plan assets totaled $3 billion, and the company's pension benefit obligation was $892 million.
This is not a new phenomenon, as the pension plan has been in an overfunded status for many years. A reminder about our pension history: the company has not made any meaningful cash contribution to its pension funds since the 1980s, which means that for several decades, the company's pension benefits and related costs have been fully funded by the pension assets and, importantly, not out of the corporate treasury. So what has the company done with the pension plan over the last few years? First, we have implemented multiple separation incentive programs and voluntary retirement incentive programs across our various businesses. These programs were put in place to reduce ongoing costs at certain business units, leveraging the assets of the plan to fund early retirement and severance benefits related to headcount reductions.
From 2020 through the end of the third quarter of 2024, $49 million in early retirement and severance costs have been funded by the assets of the pension plan at Kaplan, Graham Media Group, Corporate, Dekko and Forney, as well as at Code3, Slate, Decile, Framebridge, and the former Leaf businesses. Providing separation benefits in the form of non-cash pension credits allows us to provide meaningful benefits while also being mindful of the business needs of our operating companies. Second, the company modified certain elements of the pension plan as follows. As of January 1, 2024, we increased pension benefits for many current plan participants. This generally ranges from a 4%-10% annual salary contribution amount, depending on business unit and years of service, subject to various plan limitations.
A new defined benefit pension offering was added for employees of Framebridge and the former Leaf businesses at the beginning of the year, and as of July 1, Hoover joined as well. The company provides education and online access to plan participants relating to the growing value of their pension benefits. Pension contributions and interest credits are now posted monthly to participants' accounts with online access and detailed transaction history alongside their 401(k) balances, all of which is now integrated into our retirement portal, providing participants with a holistic view of their consolidated retirement benefits. Finally, in looking for ways to offer pension benefits as a recruiting and retention tool, you may recall that we rolled out the pension retention credit program at Graham Healthcare Group in 2022.
With nursing and clinical field staff shortages across GHG's operations at the time, this pilot program offered a pension-funded retention credit of up to $50,000 per employee, cliff-vested after three years of continuous employment. Due to changes in the labor environment, GHG is no longer offering this program to new employees, effective April 1, 2024. Additionally, GHG recently announced increased pension benefits to most of its employees effective January 1, 2025, with a higher annual salary contribution amount. Overall, the GHG pilot program was successful. We learned a lot about how to plan, analyze, and track benefits and costs from this innovative program, and we will explore opportunities to implement additional programs like this in the future to the extent there is a strong business case.
These increased pension benefits were partially offset by the elimination of company matching contributions for certain defined contribution benefit plans, which were previously funded from our corporate treasury. The company's 401(k) plans also continue to be an important retirement savings vehicle for our employees, with high levels of participation and automatic enrollment for the majority of new eligible employees. Our strategic shift to enhance and increase pension-funded retirement benefits while maintaining 401(k) plans provides employees with compelling retirement benefits and retirement savings opportunities and minimizes retirement benefits funding out of the corporate treasury. Overall, pension service costs for the five-year period from 2020 through the end of fiscal 2024 is estimated at $170 million. While pension service costs is an operating expense, it is a non-cash charge, as these benefits are not funded from operations, but rather from the overfunded pension plan.
As a result, the company's operating cash flows benefit significantly from increased amounts of non-cash pension expense. Other more recent activities include the company's closing on a large group annuity pension de-risking transaction in October 2024. This was similar to an annuity transaction we completed in the fourth quarter of 2019. This transaction involved the purchase of an irrevocable group annuity contract from an insurance company for $461 million, which was paid from pension assets, to settle $458 million of outstanding pension obligations to certain retirees and beneficiaries. We believe this group annuity transaction with a highly rated insurance company is good for both retirees and for the company. We are proud to continue to honor our commitment to retirees and beneficiaries of the plan.
The insurance company is now required to pay and administer retirement benefits owed to approximately 1,850 retirees and their beneficiaries, with no change to the amount, timing, or form of monthly retirement benefit payments. With recent favorable pension asset returns and continued elevated interest rates in 2024, we believe the timing of the transaction was beneficial for the company as well, and the pension plan funded ratio has now improved to an estimated 6.5 times overfunded as of October 2024. As a result, we have increased overall flexibility for the pension plan, and we've created opportunities for improved efficiency and a reduction in related costs to administer the plan. As a result of the October 24 pension transaction, we estimate that we will record a non-cash pre-tax settlement gain of approximately $650 million in the fourth quarter of 2024.
This accounting gain results from recognizing in earnings a portion of the unrealized pension gain currently included on our balance sheet. So where does this leave us in terms of future planning for utilizing the overfunded pension plan? First and foremost, the company's pension plan will continue to fund meaningful pension retirement benefits to employees. The company expects to continue doing so for many years to come without utilizing any funds from the company's corporate treasury. Pension funding for early retirement and separation incentive programs at our business units will also continue to be available as needed. There are several other potential opportunities for utilization of the overfunded pension plan. The company will continue to explore M&A transactions that include assuming overfunded pension plan liabilities, as we did in 2017 in connection with the acquisition of two television stations.
As we have done in the last few years, we will look to provide pension-funded retirement benefits to employees from future acquisitions, shifting away from corporate-funded 401(k) benefits. This is an M&A synergy that is unique to Graham Holdings because of our overfunded pension plan. The management team will also continue to pursue other opportunities and initiatives to responsibly use the overfunded pension plan for the benefit of employees and the company. A few years back, the company reported studying the possibility of a transaction in which a portion of the existing pension plan is spun off and then terminated in order to fund what is known as a qualified replacement plan or a QRP, which could allow the company to utilize its pension surplus to fund and expand defined contribution benefits in a tax-efficient manner.
Because of the regulatory issues associated with this type of transaction, the company instead has focused its recent efforts on enhancing and expanding pension benefits for its employees. While we are not actively considered this type of QRP transaction, we will continue to monitor possible regulatory changes in this area. The company's historical management of the pension plan and related investment strategies over many decades has put the pension plan funding status and the company in a highly enviable position. At current levels, the company has both the ability to fully deliver on commitments to our current and future retirees and significant flexibility to design and fund future retirement benefits that will continue to attract and retain employees that contribute long-term value to the company for many years to come. Back to you, Tim, for closing remarks.
All right. Thanks, Wallace.
That wraps up the prepared remarks portion of today's event. At this time, we'll move to the Q&A session. Thanks to those that sent in some questions in advance. As mentioned earlier, you can put in questions in the presentation portal, and we'll go through them for as long as time allows. With that, we'll start off with a couple of the ones that came in advance. We've noticed M&A activity slowing this year. I know you can't discuss anything concrete in the deal pipeline. Are you finding anything of interest, or is there simply plenty of profitable growth fodder internally for the foreseeable future? We have had opportunities both via share repurchases and to fund some organic initiatives, both at Healthcare and a few of the other businesses that we felt really good about.
That has been a place where we put capital. That said, we continue to turn over rocks, and we turn over a lot of rocks, and most of the time, there's nothing underneath them. But I suspect at some point in time, we'll turn over a rock, and there'll be something good. So I do think that M&A has been part of the DNA of the company and will continue to be. But in the recent years, whether it's share repurchases or funding some organic opportunities, that's one in the capital allocation calculation. Next question is, CapEx is down year over year despite the revenue growth. Assuming this is just a timing thing, but do let us know if this is part of the greater spending discipline.
I think CapEx has maybe been slightly elevated the last few years as we've had a few growth CapEx opportunities, whether it's been rolling out stores at Framebridge. There have been a couple of, at the manufacturing facilities, a couple of new plant construction projects that have occurred. I think that we've seen maybe greater growth CapEx opportunities in the last year or two than we've seen historically. Absent those growth CapEx opportunities, CapEx is probably likely to actually come down a little bit. If there are new opportunities that we feel really good about the return profile, we'll obviously pursue those as well. I would say on balance, it's probably been slightly elevated because of more growth opportunities and actually is with absent future opportunities, would likely be down a little bit on a steady-state basis.
You state that CSI is the driver, which continues very nicely along the trends you discussed at the annual meeting. Any plans to move even faster in the general area? It appears that you've struck gold with this relatively new segment and curious to see if there is any step on the gas type of thinking. Look, we've been doing a lot in the healthcare segment, and I talked about it a little bit in the remarks here, and we've invested and continue to this year in building out an infrastructure to build a bigger segment and operation there. We continue to invest in that space, although the revenue growth is starting to show that operating leverage in that segment beyond the investments that we've made there. Look, I think we're open to finding good opportunities to go faster.
We haven't found them yet beyond our internal opportunities, and we have been growing at a pretty good clip. But if there are things that come up and make sense, we're not foreclosed from that possibility. All right. The next question is, the strength in KI, Kaplan International, seems pretty broad-based. Any one or two areas that really excite Andy or yourself over the next few years? I might actually push this over to Andy, as he's probably in the best specific position to go and answer that. An dy?
Hey, Tim. Actually, I think you said it very articulately in your opening comments. We feel really good about the position of Kaplan International. Assets and capabilities against what we think are going to be really favorable macro trends over the next 10 years or so. That's what you were just talking about.
In particular, we think that KI student sourcing, reaching capabilities are areas that are very strong right now, and we keep on building them out. And I think we've got an array of highly desirable university partners and institutions and our own programs that we think are unparalleled, and we keep on adding to those as well. And so I think those of you who have been following Kaplan and really Graham Holdings for a long time know that we don't sort of wave a flag and say, "Look over here. Look over here. This is the sexy new element." It is continuing to build out a strategy that we think is going to serve our shareholders in the long term. And we feel really good about the fact that we've positioned our portfolio in an area that we think is going to grow over the long time.
Having said that, to be clear, international education is subject to periodic disruptions as destination countries adjust their student visa levels or their economic cycles in individual countries or geopolitical events and so on. And we will see those things from time to time. In fact, Australia has got an election coming, and one of the elements that has become an issue there is they're pulling back on openness to student visas. And that has an impact on all the universities and those who serve those universities. But we've been at this a long time, and these things are cyclical, and we're very optimistic about the long-term picture of KI.
Okay. Thanks, Andy. All right. Next question. I'll turn over for Wallace to answer. Other than what's been presented today, are there other plans regarding leveraging the rest of the overfunded pension?
Is there anything further about the impact of the annuity deal in the pension plan? I'm assuming this is totally different from some of the previously contemplated solutions to untrapping the surplus. Is this a creative solution to that ongoing puzzle versus liability management? Wallace, why don't you take that one?
Okay. So the annuity transaction that I talked about that closed in October, it was a pension de-risking transaction similar to a transaction that the company closed in Q4 of 2019. So this did not result in any so-called monetization or anything like that. There was no reversion of assets to the company associated with the transaction.
In simple terms, we talked about the company, if they engaged in a pension plan termination with a cash reversion to the company of the surplus, we would be faced with a 50% federal excise tax on the reversion amount, plus federal income taxes, plus state income taxes, among other costs. So this is not attractive to us given the significant adverse tax consequences. There was a recent transaction by another company that was recently announced, and that company announced a pure plan termination with a QRP replacement. They have very different tax attributes in terms of current taxes that they would be required to pay. The transaction that we were talking about, the QRP transaction that we had previously been considering, was not a pure plan termination with a reversion.
It was, as I talked about earlier, a transaction in which a portion of the existing pension plan was spun off and then terminated to fund a QRP. So more complex. And as I mentioned, there are regulatory issues associated with that type of transaction currently that we're monitoring, but at this time, we're not actively considering that type of transaction. The annuity transaction puts the pension plan at a higher funding ratio, which I talked about, but very close to the same overall surplus level. It gives the company more flexibility to leverage pension assets in the future. And as I mentioned earlier, we look at the overfunded plan as a significant long-term value for the company, and it'll be managed for the long-term benefit of employees, future employees, retirees, and the company.
Okay. Thank you, Wallace.
Next question is, the company's tone has been that there will be gradual retransmission declines due to cord cutting. Any change in your outlook and any traction yet with the digital local initiatives, financially speaking? So maybe I'll take the first part of that, and then I'll ask Catherine to chime in on progress on digital and local initiatives. No, I think our outlook remains pretty similar, sort of what I discussed in the remarks, which we've seen kind of mid to high single-digit subscriber declines on the traditional side, and then with some offset on the virtual MVPD side. I think that is what our expectation is for the foreseeable future. And Catherine, do you want to speak to any progress on kind of newer digital and local initiatives that you've been working on and talked about in the past?
Thanks, Tim.
So as you know, we've successfully built and continue to build a substantial digital audience across all of the platforms. And when we look specifically at streaming, we're seeing significant revenue potential emerge as a real sustainable business opportunity. We know that when we look at the value of streaming plus what we can do on linear, it perfectly pairs together. It really allows advertisers to build comprehensive campaigns that reach viewers wherever they consume content, which is one of the reasons why we continue to invest heavily in making sure that we are reaching the audience with the content they want where they want it.
We also think that we're going to continue to see revenue growth through things like local sports and events, and that working and continuing to create exclusive content that resonates with our communities is going to really allow us to open up even more sponsorship opportunities in ways we've never seen before. So it's not just really also about the future revenue potential. We're actually seeing meaningful traction as our advertisers are recognizing the power of reaching cord cutters and a digital-first audience through our own trusted local platforms and beyond. And we really believe that the combination of both our linear strength and our incredible reach plus growing digital presence positions us very uniquely in the market where no one else can compete with us.
Thanks, Catherine. All right.
The next question is, year to date, you've had good year-to-date revenue growth and strong cost controls across the board, which has been translating into operating growth with revenue growth growing and translating into great overall cash flow growth. Tell us more about the firm-wide operating leverage we're seeing in 2024 and its sustainability. We have a lot of different businesses that have different margin structures, different growth rates. So it's hard to have an overall company-wide view or perspective on that question. What I would say is that our broadcast business, our Kaplan business, and our healthcare units are some of the more scaled businesses that we have. When scaled businesses grow top-line revenue, you're normally going to see more operating leverage come through than some businesses that are less scaled. So you're probably seeing that a bit more in this year.
But I would hesitate to say it's a company-wide or firm-wide operating leverage initiative just because the different units are so different in their structures overall. Next question. I'll have Wallace take this. The Purdue receivable continues to be large. Is it something that is just a fact of life, or do we get a balloon payment at some point? Wallace?
So K aplan had a receivable of about $114 million at September 30, 2024, from Purdue Global, which we disclose on a quarterly basis. So first, there is no balloon payment related to this. There are four components of the balance. First, an amount due for service reimbursements. Second, estimated fees payable to Kaplan for the current Purdue fiscal year, which is a June 30 fiscal year. Third, a deferred fee amount carried from prior periods.
Then the last piece is a long-term receivable of roughly $20 million related to the advance as part of the initial KU transaction back in 2018. This balance is relatively consistent over the last three quarters, and we don't anticipate any large changes, at least in the near term. It was $98 million at the end of 2023, $106 million at the end of the first quarter, $114 million at the end of June. Also, we have confidence about the collectibility of the receivable.
Okay. A couple of questions came in referencing the Eastman Kodak pension QRP transactions. So maybe I'll bucket them together. Wallace touched on that quite a bit. But essentially, the questions, if I meld them together, are a version of, does it make it more appealing for Graham to pursue a path that's similar?
I guess what I would say is, I'm not an expert on Eastman Kodak's pension plan, and we've looked at the transaction and seen it. I guess the only thing I would add beyond what Wallace said is that we're in a pretty different scenario in the sense that we have an active plan that we, as he referenced, that we use to provide benefits in lieu of 401(k) matches, to provide benefits in lieu of severance that gets paid out of treasury. Just our overall calculation is a little different. But it's always helpful when you see other data points that are out there to kind of push your thinking and see what may or may not make sense for the company overall. Okay. Next question.
Given the potential changes in the broadcasting regulatory environment in a new administration, do you perceive Graham Media Group as either a consolidator or a seller over the next four years? Well, I think we need to. It's hard to totally project what may or may not happen before administration has taken place and before new policies have been put forward. I guess what I would say is tying back to the remarks in the commentary I gave, we never really thought or saw as much merit in the national consolidation for local stations. If I went back and looked at various different investment banker decks that were sent to me in 2017, talked about retrans and subs growing to the moon. It talked about extracting more synergies in negotiating relationships, and I actually was doing this not too long ago.
I looked at some of the projections from those decks, which was the primary reason for consolidation in the space at that point in time of what they thought net retrans revenue for the industry would be, and then I looked at what it is actually today, and it's a laughable gap, so I think that we didn't really buy into that theory. I think that has largely borne out in a way that we feel okay about and the choices we've made over that period of time. We do think there are benefits to local scale, that if you have a local footprint that is larger, if you have the ability to reach more people, to let resources be leveraged further in a market, that that can have benefit financially, and it can have benefits in the quality of the product and what you can invest in the product.
S o to the extent that, as I referenced, the Top-Four Rule were to change or augment in some ways, and we have some views on how maybe better or worse ways that could happen, I suspect that we would look at consolidation in a different way than we have previously. The specifics are very TBD depending on what actually happens in that world, but we do think there are benefits to local market-level scale in a way that is different than the benefits to national scale. And as such, our perspective on it would probably be different. All right. Next question is, how does Graham Holdings plan to integrate AI into its education or media segments, and what impact do you anticipate over the next five years? Well, this is a question I really, it's a real-time evolution here.
Maybe I'll start with a little bit of a high-level view of how we view it, and then maybe I'll push over to Andy, who's, as I referenced in the remarks, Kaplan's spending a lot of time here, and he can chime in as well. So at a high level, we think about AI in really two different buckets. One is what I would call sort of inventive AI, so new things that didn't exist before that AI has allowed to happen. And then the second is more on the productivity side, and your resources can go further than they used to. I would say we think a lot of the gain in the short to midterm is probably on that productivity side, and that's where not all, but a lot of the focus has been, whether it's at Kaplan or the Media Group or other businesses that we have.
But there is effort going in on the inventive side, and I would say maybe Kaplan is furthest ahead in some of that thinking. So maybe, Andy, I'll pass it over to you to chime in with a few examples of some of the things we're working on.
Yeah. I mean, I think if I heard the question correctly, it asks how we plan to integrate AI, and I would say that our existing students and partners are very much getting the benefit of AI innovations already. And that is in the experience that they're getting, so augmented offerings with personalized learning experiences and AI tutoring, grading, and assistance, and so on, that help students improve both their engagement and their outcomes. I mean, that is very much part of our programs already.
As you say, Tim, we use AI to make our business more efficient and more effective in our marketing and our call centers and so on. I think if you think about over a five-year period, one thing we could be pretty certain of is that students are going to be much better off. It would be hard to sustain an academic program that doesn't have 24/7 tutoring and an unlimited number of practice questions and things like that. Students are going to be getting a lot more and more tailored as well as more content that's going to be available to them. That's great. That's great for everybody. I think for an investor, you'd say, who is likely to be a winner in that world, and who's most likely to be a loser in that world?
Because some organizations are going to be better able to provide the kinds of experiences that may become table stakes and to invent new experiences that were not cost-effective to deliver previously. And the way we think about it, the vast bulk of money that's spent in education is spent through public funds, mostly in public entities that provide the delivery. And it just seems to me more likely that some of the private sector providers like Kaplan are going to be more nimble and more effective at leveraging AI over time, such that dollars that are currently spent in the public sector are more likely to start to shift over towards the private sector as that happens.
Obviously, Kaplan has no entitlement to any of that money, but if we do our jobs right and we're very focused on it, I think there can be a real opportunity for us there. So we're very cognizant of the risks that AI presents, but we are also really, and I would say much more so, very excited about the opportunities. We're working hard to guard against the risks, and we're really excited about how we can provide much better experiences for students and kinds of experiences that haven't existed previously and get paid for it. So, Tim, you know me. I can keep on going, but I think.
No, that's great, Andy. I appreciate it. All right. Next question is curious why you have not sold your TV station group over the years. Is it because the tax base is so low? You don't sound very positive on the industry.
So, there's always a temptation to view something as, are you a buyer or are you a seller? And oftentimes, you can just be somebody that holds, and that can be the right answer. And I think that that is really how we've viewed the world. We've had a business that we understood with a management team that we liked that's been producing a lot of cash that we could use for other purposes. And when we saw the opportunity set that was out there, nothing ever has changed our perspective on that, that was the right thing to do. Obviously, having a low tax base is something that would matter in that circumstance. Touching back on my previous answer to the question around consolidation, we never really saw the benefits of the national consolidation play.
We think if there's a world where local consolidation is more prevalent, it is, we'll analyze and look at, is that something that makes sense, probably in a different way than we traditionally or historically have looked at on the national side. And that could be as what way we would look at and participate in that depends on kind of what the opportunity set is. So, okay. Next question is, nice to see the annuity transaction in October. Is it possible you will annuitize even more of your pension obligations? The short answer is yes, probably not in the near term. This is the second annuitization transaction that we have done. As Wallace referenced, we did one, I believe it was 2019, could have been 2018, but about five, six years ago.
If there's a right set of circumstances that existed in the future where the combination of what we could do, what the assets looked like, what the liabilities were, and sort of discount rates were where we thought everything lined up, we would look at doing that again. With what we just conducted, it's unlikely that one would happen in the short term again. The next one is, what are your plans for the outstanding bond due in 2026? Wallace, maybe I'll push that one over to you.
Sure. Overall, we're comfortable with the company's balance sheet with cash and marketable equity securities that exceed our total debt. Regarding the eight-year notes that are due June 1st, 2026, we're out of the period where we would have a penalty, so we've certainly been monitoring and looking at the markets. We'll look at that in 2025.
We'll assess that, consider what we should do, when we should do it. But it will depend on the company's overall position at the time. It'll depend on what interest rates look like and other factors. I would say that overall, we do believe a blend of fixed and variable debt makes sense overall for the company.
Okay. Thank you, Wallace. Can you speak to the continued losses in other businesses, and what would the cost be to close, restructure some of these businesses? Hopefully, I addressed some of that in the remarks. I would say there are some losses in there that we feel, there's some profits in there in some of the units, but there's some losses in there that we feel very good about in terms of the investment that we're putting against something relative to our expectation of return profile.
Those, I would say, we're not in a spot where we're evaluating close or restructure. On most of—just in general, on most of those businesses, there's not a—if we were to make a different decision on something, it's not an excessive type of circumstance in terms of cost. In particular, we have the pension, and we have done SIPs, as referenced before, that could be leveraged to kind of protect the treasury as part of any kind of wind-down or restructuring that we would look at in a hypothetical business. But for the most part, I don't think there are extreme structural issues that would cause very expensive things to occur in a wind-down scenario because a reasonable number of the businesses are on the smaller side, and so they would only kind of take so much in order to wind down. All right.
A couple of Framebridge questions that I'll combine as one. So how close is Framebridge to profitability if one excludes store opening expenses from the calculation? And given that it will operate over 30 stores at the end of the year, how are we seeing the economics of the business evolve as it continues to scale? And what will we think it will take until Framebridge reaches break-even? So I guess in the first part of that, how close is it to profitability once it's including store opening expenses? We haven't broken that out specifically. It's closer than it used to be is probably how I would answer that. What are we seeing from an operating and economic standpoint as we get more scale?
As we open more stores, what we're able, there's sort of scale that we get on our, what I would say more of our direct costs in terms of cost of materials, shipping, cost of, from an overall marketing standpoint, when we have more return customers, we're not spending quite as much on a new customer acquisition side. And then on the indirect side, as we get greater scale overall, we see the ability to put more units through our existing studios and production facilities. You really get a lot of operating leverage on that. And so as we have been able to grow the business from a store account standpoint, we see those come through more and more. I think we have more to go. I believe that the overall margin structure, if this were to become a much bigger business, will still be substantially improved from where it is today.
It's relatively clear when we look at the business that when you achieve more scale and more unit volume, that the unit-level economics will improve as that scale materializes. I think the question around how long will it take until Framebridge reaches break-even, it's really a bit of a question of we probably could make that happen sooner, and it would be the wrong answer. If we look at the level of kind of opportunity from a store standpoint, and we look at the potential manufacturing studios and production studios that we would create to service those, it's a little bit of a hub-and-spoke model. When we would kind of build or open up a new area, you invest some from a pre-opening standpoint, but the unit economics of that spot are actually, from everything we've seen, quite strong. You have some CapEx to do that.
You have some startup costs, and you've got to get the unit volume up there. And when we look at the kind of broad-based opportunity and say, "Okay, could a brand have 20% of this market in the long term?" that feels very achievable to us and something that could happen. And so I think that if we were trying to become a profitable business as soon as we could, it would probably be at the expense of achieving that 20% type of share that we think is possible. And so it's really a balance based on what are we seeing in the business and the unit economic profile that drive those decisions. So I hope that's helpful because to sum it up, we could be profitable sooner in that business, but we also think there's a good chance that that would be the wrong answer. All right.
Please talk about potential cash accumulation over the next few years and use of the cash absent acquisitions. Thank you. We expect the business to. It has produced reasonable amounts of cash over the last few years, and we are hopeful and believe that there's a good possibility that that will continue. I guess the question, it feels hard to answer in a vacuum because it really depends on what the opportunity sets are. Certainly, we think there are some internal opportunities that we continue to fund on that front, probably not at the level that would match cash accumulation, but we're fine. The right answer were to pay down debt with cash, we'd rather do that than have it sit there. We're fine with waiting for the right opportunity to come across if that were to be.
But it's really a question of what does the opportunity set look like as we're building up a cash balance? What's the share price lo ok like? Are repurchases something that would make sense at that point in time if we couldn't find acquisitions and we're fully funding internal operations? So I think we really look at it as a capital allocation choice basis, and if M&A were not one of those choices, we would look at returning capital to shareholders, paying down debt, and funding internal acquisitions - or sorry, internal opportunities based on and kind of evaluate based on those criteria. We've had a couple of nice wins in test prep as of late. Is the institutional product offering starting to really resonate? Andy, why don't you take that one?
Yeah.
As an observant shareholder, yeah, we have had some good wins, and I think that it's starting to resonate maybe a little slower than I'd like, but we've introduced a suite of all-access programs which take advantage of Kaplan's breadth to universities and to companies and to governments. And people are really starting to understand how they can create an effective program for their constituency by saying, "Whatever your needs are educationally, Kaplan can help." And we like that as a means whereby an institutional payer is funding as opposed to a bunch of individual consumers. So yeah, it's resonating, and I think we hope you'll see more of that in the future.
Okay. Well, that answers another question which just came in too, which is what needs to happen to return Kaplan's supplemental education to revenue growth?
I would say, unless there's anything else you want to chime in on that, Andy, some of the things that you just talked about the institutional side are really part of that equation. Anything else you'd add there?
Yeah. I mean, we're pretty much stable on the revenue side in supplemental education, but growing on the income side. And we think as we shift the business towards a little bit less focused on consumer, although we think we're seeing good progress there, and more towards institutional, that is going to go to our strengths, and I think that'll be good for Kaplan.
Okay. Next question is, it looks like GHC has acquired additional shares in PubMatic in 2024. What does GHC see in PubMatic? I don't really have a policy on talking about individual securities that we would acquire.
We added very modestly to it over 2024, and it's a pretty small percentage of overall assets, and beyond that, I don't think it serves us well to discuss the specific securities, so. Can you talk about any industry trends in the in-home healthcare industry that you are particularly excited about? What is Graham Healthcare positioning itself to capitalize? What is it doing to position itself to capitalize on those tailwinds? We like the in-home care industry because we really think it makes sense for society. If you look at the businesses we operate in, patients would almost always rather be treated at home. Their satisfaction is higher. The outcomes tend to be as good or better than in an acute care facility, and the payers, which in our case is not always, but largely Medicare, have reduced costs relative to providing the care in an acute care facility.
If it's better for the patient and it's better for the payers and the acute care facilities would rather not administer a lot of this care in their facilities, that feels like a pretty good dynamic and place to be in the industry. I would say the second thing that we like about it is it should be a market demographically that continues to grow. Similar to how we referenced earlier, you can look at the number of graduating high school seniors there will be 10 years from now by the number of second graders today. You can look at the number of people who are likely to need in-home care services, particularly on the home health and hospice side, based on current age demographics and cohorts today. We like what those trends look like for the foreseeable future.
So I would say that before we even talk about our individual business, we think this makes sense for society. It's part of reducing the overall healthcare costs with good outcomes, and we think it's a market that is likely to continue to grow in terms of its needs for service. I think with our business, we have a really mission-oriented group of folks that lead our team and figure out how to go and excel in providing care and excel in innovating in the provision of that care and figuring out how to do it really effectively. And when you do that, you're able to grow organically because you're providing great service to the patients and the providers that you're working with. And when that happens, you get a formula that can build on itself, and that's really what we've seen.
And so we're continuing to focus on providing care, making sure that we have the operating capabilities. And I know I've referenced over the last few years, we really invested a lot in the infrastructure of Graham Healthcare Group to become a bigger business. I think you're now seeing that that bigger business has been materializing. And it's not revolutionary, but I think we'll kind of continue the formula that we've been doing for the last few years over the next few years, and we feel pretty good about that. Okay. The next question is, in the past, I recall some optimism on manufacturing, but that doesn't appear to be happening. Please detail your view on how to improve the businesses therein. So there are four businesses that live within our manufacturing segment.
I would say that two of the businesses have been very, very steady and in line with historical results or even doing maybe incrementally better than historical results. The biggest business within the segment is Hoover. We really like Hoover and its product and its position in the industry. We like that quite a bit. It is a business that can have some cyclicality in it. For the first six or seven years of ownership of Hoover, we saw kind of mid-cycle or above-cycle parts of it. We've seen for the last probably 12 to 15 or 18 months the down-cycle part of that. And while we would love for there never to be a down-cycle, we know that that's unrealistic. We don't think there's anything wrong with that business or that industry. It's very much tied to multifamily housing as its largest end market.
And if you look at permits on multifamily housing, if you look at projects that have been created, it's down from where it was before. And in fact, we know of projects where people are waiting for an interest rate environment to make the economics of a project work and make sense. And if that were to happen, we would start to see that business recover some. So I would say that the last good thing is that the profitability at Hoover remains very, very solid even when it's in a down market. And we really like that business, and we expect over the next decade that Hoover will produce a lot of returns for shareholders.
The second business has been Dekko, and Dekko's largest end market was commercial real estate, which has been, I think, no surprise to people, has been very challenged over the last four or five years. I guess two comments I would make. One, Dekko had supply chain challenges in addition to its end market challenges where it bought ahead some inventory to try and protect customer relationships kind of at the peak and in early days to peak of pandemic. Some of what you're seeing from a results standpoint is us kind of working through that inventory either at reduced margins or in some cases writing off a little bit of that inventory. So you're seeing some of that come through, and that is a relatively short-term piece.
The second thing I would say about Dekko is we are actually seeing some green shoots in that business for the first time in a handful of years where we feel optimistic that it could be on a path to growing again as we've kind of worked through the worst part of the commercial real estate cycle. So in summary, we like the segment. We believe that it's going to be a good, strong contributor. We do have two businesses, including the largest one in a down cycle right now, but nothing kind of fundamental about Hoover or the segment in general makes us worried about its future at the company. Okay. Questions. You mentioned earlier that Graham pursues repurchases when shares trade at material discounts to intrinsic value. Shares are up $300 since last year's investor day.
What do you think about Graham's intrinsic value per share today versus the company's share price? I have not and am unlikely to ever put out a view of what my view of intrinsic value will be. Certainly, a $300 per share rise is substantial relative to the starting share price, and that gap has closed. So I think that gap has closed from what it was. I don't want to opine on how much of a gap may still exist, and I don't want to opine on what my view of the exact price would be today. But certainly, when it's gone up 40%-50% in a year, I don't think the intrinsic value has grown 40%-50% over that same period of time. So that inherently implies that the gap would have closed somewhat in that period.
With that, we are both bumping up on time and bumping up on the end of the questions people have submitted. I appreciate everybody's time, and thank you for your participation in the 2024 Graham Holdings Investor Day. Thank you.