Graham Holdings Company (GHC) Earnings Call Transcript & Summary

Good day, ladies and gentlemen, and thank you for standing by. Welcome to today's Graham Holdings Company Investor Day webcast. A quick reminder that today's program is being recorded. And at this time, I'd like to turn the presentation over to Timothy O'Shaughnessy, Chief Executive Officer. Please go ahead, sir.

Great. Thank you very much. Welcome to the Graham Holdings Investor Day. Something new we're trying out in 2020. We wanted to provide an additional opportunity outside of the annual meeting to update you on how we view the business and allow you to ask questions to help you better understand our company. Thanks to all who have joined today as well as those who have submitted questions in advance. We have a set of presentations that several members of the senior management team and I will walk you through. I will kick things off, and then you will hear from Andy Rosen, the CEO of Kaplan; Emily Barr, the CEO of Graham Media Group; and Jake Maas, our Senior Vice President of Planning and Development, who will cover our manufacturing businesses. I'd like to begin with a summary of Graham Holdings through Q3 of 2020. I can use words or phrases like unprecedented or enormous challenges or switch to remote relatively smoothly, but it is likely you've already heard these stories from other businesses. All are equally the case for Graham Holdings, but they are unlikely to provide you with any truly helpful information. So I'd like to call out a few other words that describe our year. Pivotal. Several large capital allocation decisions, paired with accelerated digital transformations at Kaplan set the stage for value creation over the coming years. Inspiring. Our small corporate team punched way above its weight in 2020. Members of our team and our managers have never worked longer hours for lower compensation to drive results that we've never been prouder of, odd since we would have considered these results disappointing on January 1. Suffice it to say, I could not be more proud to work with the business leaders at Graham Holdings. And I appreciate our managers for what they've done to keep our businesses healthy and prosperous in 2020. Let's move to our financial results. Through Q3, our revenue is down 3% from 2019, and our adjusted operating income is down 18% to $127 million. The business performed very well in the early months of the year before the sudden onset of tremendous pressure related to COVID-19. After bottoming out in April and May, in aggregate, our businesses have made a slow and steady recovery. We have several operations still acutely financially impacted by COVID and a host of others operating in a recessionary stance. Nevertheless, the company's net cash from operating activities through Q3 was $241 million, up from $72 million in 2019. As of Q3, our cash and marketable securities totaled $802 million against $515 million of debt. We remain somewhat guarded about what the future holds, but believe, in 2020, we have protected our underlying earning power and possess a balance sheet that allows us to pursue future opportunities. Graham Media Group had a tremendous recovery of political spending in Q3, and into Q4, drove results far better than we could have hoped. Kaplan's transition to a digital-first business accelerated and we began to see strong growth at Kaplan higher education. You'll hear more about both of those things in detail from Emily and Andy shortly. We do not have a feature presentation today from Graham Healthcare Group. However, I do want to take a moment to speak about our results because of their importance to Graham Holdings. If you haven't familiarized yourself with Graham Healthcare Group, you should. At its core, Graham Healthcare Group provides in-home nursing care in the fields of home health, hospice and infusion services. They do this in a few ways, and GAAP accounting doesn't always make it easy to connect the dots, so we'll do so here. We have several health care businesses operating primarily in Michigan, Pennsylvania, Illinois and Texas. These show up in the health care segment of our financial statements. And through Q3, those operations generated $147 million in revenue and $24 million in adjusted operating income. We also have 5 joint ventures with hospitals and other health care providers in these areas, where we own less than 50% of these entities and manage the operations. Our share of these earnings is very real, but show up within our equity method investments line of the income statement. Through Q3, our earnings from these health care joint ventures were $8 million. Overall including these joint ventures, GHG managed businesses that generated $241 million in revenue through Q3 of 2020. Our co-CEOs, David Curtis and Justin Dewitt, have done a wonderful job of improving the performance of our operations on almost every dimension, from quality of care to financial results, despite the pandemic. COVID-19 has brought a unique set of challenges to our health care business, which includes significant PPE expense to keep staff and patients safe. With more than 500 patients we care for testing positive for COVID-19 and roughly 10% of our home health and hospice workforce having missed time, mostly due to quarantine for potential exposure, the staff at Graham Healthcare Group has been exceptional at continuing to provide care to vulnerable and homebound patients. We are proud of the team at GHG and pleased with the improved results. We remain optimistic about our ability to continue to grow this business over the long term. Outside of operations, our pension plan continues to be very well funded. As of September 30, our pension plan had assets of approximately $2.5 billion, a balance which continues to far outweigh our obligations. The pension is valuable to the company in several ways. First and foremost, the plan is designed to provide a benefit to current and former employees of the company. It also provides a valuable and increasingly rare defined benefit that can generally reach about 8% of annual salary. In addition, the overfunded status of our pension plan is a very valuable feature of our financial statement. It allows us to: one, provide generous pension benefits in lieu of a higher 401(k) match. Each incremental point of 401(k) match for pension plan participants could cost the company approximately $3 million in negative cash flow annually; two, on those occasions, when it becomes necessary to reduce staff, the pension plan enables us, in some circumstances, to provide enhanced payments to transitioning employees, providing a helpful benefit in a difficult time. Over the past 6 years, these benefits to employees have averaged about $4.4 million annually, including $14 million through Q3 of 2020; three, in an acquisition, we may be able to assume the pension liabilities of the acquired company as part of the transaction. We most recently did this with our acquisition of 2 television stations in 2017. The capacity to support employees of an acquired company, through the assumption of pension liabilities, uniquely positions GHC to succeed in a competitive M&A landscape. We remain optimistic that over time, we may find additional opportunities along these lines. We continue to actively explore additional ways the overfunded pension can be helpful to Graham Holdings, and we will keep you posted along the way. Our 2 major capital allocation decisions to date in 2020 have been a substantial repurchase of shares and the pending sale of Megaphone. Through Q3, the company spent $123 million repurchasing approximately 6% of the total outstanding shares at an average price of $383 per share. As a reminder, unlike many companies, we do not have a program that repurchases a specified number or dollar amount of shares. In both my and Don's tenures, we have gone many quarters or years even, where we did not repurchase any shares. We have also had times when we repurchased several percent of the company within a single quarter, 2020 is one of those years. We repurchase our shares when several criteria are met. First, we feel confident we are able to fully fund any CapEx and operating needs from our businesses. Second, we have ample dry powder to opportunistically act should any compelling opportunities arise. And third, we believe the stock is trading at a material discount to our view of intrinsic value. We thought this true through much of 2020. All 3 of these are judgment calls, and we will take a conservative view when assessing each bucket. We'd rather miss out on repurchasing a few shares than place the company at risk or not be able to pursue an opportunity that makes sense for the business. However, when these conditions are met, as we believe they have been in 2020, we think repurchasing shares is a wonderful thing for our shareholders. I'd like to provide some color on our other major capital allocation decisions so far in 2020, our announced sale of Megaphone to Spotify for $235 million, relative to both our logic and financial returns. We started Megaphone within Slate in 2015, and we owe thanks to Slate for being an early launching platform for the business. As it became clear that the future of the business and the opportunity was in technology as opposed to content, Megaphone moved out of Slate and became a self-contained unit of Graham Holdings, funded 100% by GHC. The company became a market leader in both advertising and podcast publishing technology that much of the industry has come to rely on. While we continue to be bullish on the future of Megaphone, we are also realistic that the next phase of the company's growth may be better served with a different owner that can provide at more scale. We view the $235 million purchase price as a fair price for the business and an excellent return on our cumulative invested capital of approximately $45 million on a pretax basis. We expect to report a pretax gain in the range of $215 million to $220 million at closing, excluding working capital and subject to regulatory approval. Now I'd like to explain our logic of why we fund new businesses within Graham Holdings, such as Megaphone. As the costs and barriers of starting a new business have decreased over the past several decades, in our view, the risk reward has correspondingly changed. We believe we have a core competency of working with early-stage technology companies, particularly those that are in related or adjacent markets to our existing businesses, to ascertain product market fit and effective business models and will, over time, create substantial value at Graham Holdings with a subset of these operations. We view them as relatively modest bets that have the ability to produce outsized returns. All things being equal, we would rather fund an organic initiative due to tax advantages. An internally funded initiative is funded with pretax dollars and reduces our internal income by the amount of the funding. An external investment is funded with after-tax dollars. Said another way, our dollars go further with internal investments, even if they reduce reported income in a way, external investments would not. In a few moments, other business leaders will do a deeper dive on our broadcast, education and manufacturing businesses. But before we go there, I'd like to provide some insights into our other businesses, which includes the aforementioned Megaphone as well as a host of other small to midsized operations. A question I've received regularly is, why does GHC own these businesses? Well, much of the world thinks in terms of verticalized industries. We do not. We make our decisions based on a set of horizontal principles, which I regularly reference in our annual letter. More directly, we own these businesses because we feel they will provide good cash-on-cash returns and grow intrinsic value per share for our shareholders. I want to highlight a subset of these businesses that are in very different sectors, but have similar key characteristics and potential future growth opportunities. Since January of 2019, we have acquired several auto dealerships, Clyde's Restaurant Group and Framebridge. One might reasonably ask what do these businesses have in common? It may not be obvious at the outset, but let me explain how we view them. They all operate in industries with track records of industry profitability, operate in extremely fragmented industries, historically have been consistently profitable or in the case of Framebridge, have profitable unit economics, and most importantly and lastly, have very, very strong management teams. In each of these sectors, we think underlying technology trends will create winners and losers that result in aggregation of market share and superior long-term economics to those who gain that share. At Clyde's, the dealerships and Framebridge, we are pursuing strategies and investing in technology that create omnichannel experiences, drive more scale and position us for superior long-term outcomes. Here are a few examples. At our automotive dealerships, we have focused on building the dealership that caters to the future needs of our customers, where they come to a dealership only if they want to; where a customer doesn't have to choose between service, convenience and price; where a dealership footprint area isn't just your immediate geographic area, because technology can cover distances quite a bit better than people can. Our new Jeep dealership, located in Bethesda, Maryland, has been built from the ground up under this model. Customers can come to the dealership in person or set up a video call. Cars can be delivered for test drives to your door, or you can have a non-commission representative sit next to you and explain the vehicle, and paperwork can largely be completed online. With this improved experience, a customer could test drive multiple cars and complete a purchase without ever going to the dealership if they so choose. So far, the signs are good. We have quickly built up market share in the area, and currently harbor a 4.8 out of 5 rating from Google reviews. As we continue to hone our formula, we will look to apply this model elsewhere. Also, on the automotive front, we have launched carcaretogo.com, a marketplace for automotive service. While early, we aim to create a one-stop shop for service needs via a valet model that does not require the customer to ever go to a service center. The vehicle is picked up and returned once the service has been completed. At Clyde's, we are employing technology in a few key ways. First, in our operations, we are using software that helps our teams improve productivity, per item profitability and menu mix, improving the overall margin profile of the business. Second, we believe consumer preference for takeout and delivery, while accelerated by the pandemic, is here to stay and will be something the most successful restaurants will do extraordinarily well, even in non-pandemic times. We also believe restaurants that become omnichannel and focus on providing a great dining experience on location and via delivery will have superior economic returns. Clyde's Restaurant Group is leaning heavily into this strategy. We have been able to use our importance in the D.C. market to work with delivery apps to create mutually beneficial relationships. More recently, we launched 3 new brands under the ghost kitchen concept, where we create a new brand and menu, but operate it as a delivery-only business out of our existing kitchens. We are optimistic these new concepts will drive high-margin incremental business to Clyde's. Framebridge is the third business that fits this mold. Jake will provide a more detailed overview in the manufacturing presentation, but we believe Framebridge can consolidate this legacy industry to become the way America frames. By taking framing into the 21st Century with both digital and retail offerings, we believe scale economics will create a very attractive business. Our data indicates that we should put the pedal down on Framebridge in 2021, and we plan to do so. Applying a technology layer to existing industries has resulted in a tremendous amount of value creation for those leading the charge. We believe Graham Holdings is a great place to drive this change in framing, automotive and restaurants because our patients in seeking long-term opportunities and capacity for reinvestment is unique relative to the competition in these industries. We may not get all of these rights, but we view these opportunities as excellent risk and reward trade-offs. At this time, I'd like to turn the mic over to Andy Rosen, the CEO of Kaplan. It's been a tremendous year of change at Kaplan, and he is eager to walk you through it. Andy?

Thanks, Tim. As you say, 2020 has played out in an unexpected way for Kaplan as it did for everyone else, and we took some significant hits from the pandemic. But COVID also pushed the education marketplace in our direction. We've been a pioneer and leader in online education for over 20 years. And when all programs suddenly shifted online, it turns out students were attracted to institutions that know what they're doing. Further, we may be the global education brand that is most widely known for practical, high-quality, fairly priced programs, and that's exactly where the market is moving. At a time when students are increasingly seeking work-readiness skills with their academics, that's another area right in our wheelhouse. Most importantly, Kaplan went above and beyond to take care of our students and partners in a difficult year for everyone. If anything, our brand has been enhanced by the quality of our response to the pandemic, and our academic outcomes have remained strong. We took some important strategic steps over the course of the last year that we believe will notably improve our earning power in the years ahead. So despite the specific challenges of this year, we feel good about where we are. Let me start, as I did in May, by reminding you of Kaplan's 4 priority areas. First, we help students qualify for, access and succeed in school, from high school to graduate and professional school. So here, we have programs that serve students starting in middle school and high school, some in partnership with top universities, as well as top -- as test prep courses to get into college, grad school and beyond and programs to help students succeed while they're in school. Second, we help students and working professionals qualify for and succeed in jobs. Kaplan prepares hundreds of thousands of lawyers, doctors, nurses, accountants, business leaders, engineers and other professionals around the globe to a range of offerings. Third, we provide services for universities, helping them attract and serve qualified students, and ensuring that those students are successful and ready for work. So we help our partner universities identify and recruit students from around the world. We host their international campuses, we enable their online programs, and we help them prepare their students for careers. And fourth, we help companies identify and employ highly qualified candidates. And we help ensure that their employers -- their employees are equipped to succeed. This includes helping businesses tap into university curricula as well as offering Kaplan programs and licensure certification, professional development and online coaching. Kaplan, of course, has an exceptionally wide range of programs and services for students, universities and companies. To a large extent, our offerings have traditionally resided in separate units within the Kaplan portfolio. Over the last 2 years, however, we've made real progress in creating student and partner journeys that further our long-term relationships. So for example, you can imagine we have a high school student, Elizabeth, from Chicago, who comes to us for ACT prep. You can see that she might come back to us in 4 or 5 years for the GMAT, so she can go to business school. And we might then connect Elizabeth to our partner, Wake Forest University, for its online MBA program. For a Brazilian student, Carla, who tests her skills in English with a KITE exam, that's our English language assessment, and then attends one of our English language courses. Then gets college guidance to Purdue service, which provides access to one of our U.K. Pathways programs, and this leads to a university degree program. Then she might decide to study with Kaplan for the CFA, or an accountancy qualification to the government supported apprentice. For a university partner, we can offer international student recruiting, then help them build online courses, add a Pathways program and house those Pathways students through an international student residents. At some point, the university may decide to extend their presence overseas and reestablish a branch campus for them in Singapore. In essence, we can build upon our relationships by adding services from across the Kaplan portfolio. And we can do the same thing for corporate partners. We can deliver licensure prep, courses from our university partners, professional development and online coursing for employees. We might also help our partners recruit from our pool of students. These examples highlight 2 distinct advantages. First, we have an enormous global base of students who we can serve at multiple points throughout their learning journeys. Second, our deep relationships with universities and businesses enable us to create holistic learning ecosystem structures that connect all 3 segments. Each segment supports the others. Historically, it's been harder than maybe it should be to offer these journeys, because our offerings were dispersed among our 4 main business units, which operated fairly independently. Earlier this year, we restructured the business, combining our 3 U.S.-based businesses into one broader unit that we call Kaplan North America. That enabled us to focus on building long-term relationships and achieve significant cost savings. The consolidation should enable us not only to target our investments in areas most likely to lead to student outcome improvements, but also to improve the margin profile of our North American business. The new unit is overseen by Greg Marino, an outstanding long time Kaplan executive, who was previously CEO of Kaplan Higher Education. He works with what, I believe, to be the best leadership team in the industry. Since we made the change over the summer, this team has made tremendous progress. With the new structure, we can avoid redundant expenses and invest more in superior technology, marketing and educational delivery and support systems. We'll better leverage our broad student database, add new student relationships and enhance student academic success and lifetime value by serving our population repeatedly throughout their educational journeys and we'll similarly build upon our relationships with universities and businesses using our rich menu of offerings to help them excel at what they do. We undertook some additional strategic moves in 2020 as well. For example, we launched 25 new online degree programs with universities, including Purdue, Wake Forest, University of Essex and Liverpool University. We deepened our existing university relationships and added new partners such as Massey University of New Zealand, University of Newcastle in Australia, Queen Mary University in London, Birmingham University in the U.K. and Simmons University in Boston. We formalized the digital-first business model for our U.S. exam prep business. We firmly established online program offerings in our international markets in keeping with the rapid acceleration of online market acceptance. And we focused heavily on student support at Purdue Global, where retention rates are up even as enrollment is growing. These moves should set us up for long-term growth, but it will be another year before we start to see the benefits as COVID dampened student enrollment in some key areas. That's not just a 2020 issue. Our enrollment carry into 2021 is lower, and enrollments will continue to be impacted until a vaccine is broadly administered. We have 3 primary areas impacted by COVID. First, 29% of our revenue comes from students who travel from one country to another to study for language programs and higher education. International travel, of course, came to a virtual halt for most of 2020. That's likely to continue into next year. In our Pathways programs, the majority of our students were willing to conduct their studies online. However, our language programs, in English, French and German, tend to attract students who are looking to have an exciting experience in a western country for 1 month or 2, while learning the local language. Until international travel returns, that business remains largely frozen, leaving our language schools almost empty this spring. 28% of our revenue comes from students prepping for standardized tests. Because these tests are often run in large on-site testing locations, many will pause for months. With no test to study for, students postpone their preparation. For those already enrolled with Kaplan, we extended our service with no additional cost. And of course, many of our programs have historically taken place in physical classrooms. We entered the pandemic with 145 student locations around the world, all of them closed during the pandemic, although some have reopened since. Fortunately, Kaplan is deeply experienced with online learning. And when we shifted our programs online, our students were highly enthusiastic about the sophisticated and engaging digital programming we were able to offer. So what does all this mean as far as our financial results? Well, we've been able to mitigate much of the impact of the pandemic, with the exception of one line of business, the languages programs that I mentioned. Language has shifted to online delivery. But given that these students seek the social and cultural experiences of living abroad, we can only make up so much ground. The shortfall in languages represented more than the entirety of Kaplan International and Kaplan's operating income decline for 2020. More specifically, we estimate the COVID disrupted losses for our language businesses will be close to $52 million in 2020. The rest of Kaplan International performed impressively under the circumstance. We anticipate languages will make a rapid recovery when global travel returns, particularly given that we've now created a more appealing cost structure going forward. We're seeing real strength, especially given the pandemic, in our Pathway business, our direct recruitment for universities, our Singapore-based branch campuses for western universities, our global professional programs and our higher professional education programs in Australia. We have what we believe is the strongest student recruitment capabilities in the world for students interested in attending western universities, with strong trusted teams across the globe. So I very much like the profile of our international business. In the U.S., Purdue Global saw significant growth. Enrollments were up by 15%. And the student census rose to nearly 35,000, up from 28,000 at the time of the Kaplan University's sale to Purdue. Student academic performance continues to improve, including retention and graduation rates. The profitability of Purdue Global and our ability to recognize our service fee improved correspondingly. While our U.S. exam prep businesses were negatively impacted by canceled examinations, each remains a leader in this market. With the shift to delivery online, we enter 2021 with a lean, focused digital model that should serve us well. As consumer preference has moved towards digital exam prep offerings even before COVID, we've been focusing on optimizing the balance between our center-based business and our delivery -- our digital delivery models. With -- when COVID accelerated student movement towards digital preparation, we have sunsetted the majority of our center-based offerings and primarily shifted resources to our digital event prep offerings. So we won't look back on 2020 with fondness, but we can look back on it with pride. Our teams responded swiftly and decisively to the COVID crisis and in doing so, protected our students, our partners, our shareholders and one another. They undertook a broad reorganization, a cost reduction effort and a strategically focus, that in combination, should increase our earning power post pandemic. Well, one of our businesses could not escape a direct hit. Kaplan, as you can see, is a highly diversified organization in terms of the types of programs we offer, the customers we serve and the geographies in which we operate. When one business is struggling and other is likely to be stronger, and that's the case here. Outside of our language business, our units outperformed our expectations under the circumstance. The pandemic will end, and when it does, we may look back at 2020 as a year that helped set the table for a new era of growth for Kaplan. And now I will turn it over to my colleague and Broadcasting Cable 2020 Broadcaster of the Year, Emily Barr.

Thank you, Andy. I appreciate that compliment, too. Good afternoon, and thank you for the opportunity to provide you with an overview and update on Graham Media Group. As a reminder, we are comprised of 7 local media hubs, covering just under 7% of the U.S., over 8.3 million households in primarily top 50 markets, with Houston as our largest market at #8 and Roanoke, our smallest, at market #71. We operate 3 NBC affiliates in Houston, Detroit and Roanoke; 1 ABC affiliate in San Antonio; 1 CBS affiliate in Orlando; and a fully local independent station, plus a CW affiliate in Jacksonville. In addition, we own and operate Social News Desk, a SaaS-based company headquartered in Atlanta that currently serves over 2,500 television, newspaper and radio news rooms worldwide, making it easier and more efficient for them to write, edit, post and curate news on social media platforms. Also based in Detroit, Graham Digital oversees the digital, mobile and OTT strategy for all our properties and is considered a leader in the field of broadcast digital operations. 2020 has been an incredibly challenging and unusual year to say the least. While the year began with a great deal of promise, in the final 2 weeks of Q1, we slammed into COVID and reverted to a remote workforce for approximately 80% of our staff. Today, we remain largely working from home with over 70% of our employees working remotely. We intend to continue working remotely until a vaccine becomes widely available and adopted by enough of the population to make coming back to the office safe for everyone. In addition, we have certain operations that have already transitioned to a permanent remote environment. COVID forced us to rethink how we perform all our regular tasks, while still delivering daily television newscasts, digital, OTT and mobile updates. Keeping our reporters and photographers in the field safe from COVID forced us to rethink workflows, eliminate all business travel, institute the use of remote interviewing through Zoom and other platforms, and upended our regular over-the-air and digital advertising, particularly in Q2. In addition, stay-at-home orders for the general public flipped the traditional demand for morning news and saw a surge in demand for midday and afternoon news casts as people sought us out for any updates on COVID. Viewership and digital access soared in Q2 to levels we had not seen in almost a decade. While viewership has since normalized closer to pre-COVID levels, we have seen the makeup of our news audiences skew younger and the engagement among our digital audiences sustain a much higher level than pre-COVID days. Because we have 3 NBC affiliates, the loss of the Summer Olympics, combined with a difficult Q2, had a material impact on our revenue in Q3, and we delivered revenue somewhat below expectations, though still quite respectable given the circumstances. Excluding property, plant and equipment gains related to spectrum repacking, operating income before amortization increased in the first 9 months of 2020. That said, next to COVID, the single biggest impact on our revenue this year was, without a doubt, political advertising. While we saw robust spending in Q1, primarily from the Bloomberg campaign, it was in the middle of Q3 that we began to feel the full impact of political on our year. The figures shown here are through the November 3 election, and demonstrate the enormous impact political advertising has had on our business, particularly when considered over the past decade. Almost half of this advertising was placed in just one market, Detroit, where we were the beneficiaries of a hotly contested Senate seat as well as several competitive congressional seats. And of course, the presidential race in which Michigan was one of a handful of states that would determine the outcome of the election. Needless to say, we did not anticipate anything close to this amount of revenue pouring into our markets this year. As mentioned earlier, we have been experiencing steady growth in our desktop and mobile usage and have been laser-focused on expanding this end of our business for many years now. From Q1 through Q3 and largely due to the impact of COVID, we have seen well over a 100% increase in users, with almost 280 million of them accessing us via desktop, mobile or tablet in a 9-month period. Breaking this down by market, you can see that all our markets have experienced seismic growth in their digital audiences. And I am very pleased to report that each one of them is the #1 source for local news in their respective regions, beating out all other broadcast, newspaper and digital-only sites, providing local news and information. We believe this is a critical strategy to maintaining and growing GMG into the future, and we will continue to innovate and build in this space because our future depends on it. As a result of increased usage, we have experienced a 13.2% growth in digital revenue through Q3, despite a substantial decline in Q2 from traditional advertisers who pulled back from both TV and digital when the pandemic first appeared. For the first time, we also garnered some meaningful political advertising online as political ad agencies recognize the value of our local sites. When we measure our value, it is critically important to recognize the impact of local news as it is the driver of all we do, in creating and distributing local media in our markets. The sale of Time inside local news and via digital delivery accounts for roughly 50% of our total advertising revenue. While we have important affiliations with various networks, depending on the market, it is the branding and identity of our station's local news operations that drives our image, reputation and business success. COVID and the recent election only served to underscore this critical point. As the news goes, so goes the station. To my point earlier, this is why we are so focused on being the #1 source for local news digitally and why we are well positioned to succeed in the future. I wanted to share with you our top 5 categories just to demonstrate where we are today in terms of rebounding from Q2 and the COVID effect. As you can see here, automotive has begun to bounce back closer to Q1 levels. And to date, we are seeing a healthy amount of advertising from this sector in Q4. Professional services, banking, financial services and other professional services dropped significantly in Q2 and is showing some increased strength as well. The overall economy and unemployment rates will drive the growth or retraction of this category into 2021. Retail was hard hit in Q2, as you would expect, and it's also starting to rebound, though we are cautious here given the latest restrictions rolling across the country. We anticipate this could continue to be a volatile category for us in 2021. Health and fitness, along with home improvement, were hurt less, as people staying at home were a natural focus for these advertisers. We anticipate these categories will continue to grow into 2021. Now I did not show you the restaurant category, which is not a top 5 earner for us. But suffice it to say, this category was devastated by the pandemic and is still figuring out when and how to advertise. I would like to add here that all our stations implemented free promotional advertising for their local mom-and-pop restaurants to support local businesses and help them stay afloat throughout the lockdowns. A significant piece of our revenue derives from the retransmission consent revenues we received from cable, satellite and OTT providers. As cord cutting and streaming continue to increase, we have been able to grow our per subscriber rates, but we are watching the trends closely and expect retransmission revenue net of network fees to slow down in overall growth in the coming years. Revenue from OTT providers such as YouTube TV, Hulu and ATT Now just to name a few, has become a small but growing piece of the retransmission revenue pie. 2020 and COVID disrupted the live sports environment and negatively impacted advertising in major sporting events that were either curtailed, canceled or delayed. The NBA Finals, NFL football, hockey and golf were all disrupted, making it challenging for local advertisers who would otherwise have eagerly invested in these events. When football resumed in September, we did experience a resurgence in interest, indicating that there remains a huge appetite for both viewing and advertising within live sports. Additionally, the cancellation of live -- I'm sorry. Apologize. Additionally, the cancellation of live sports likely contributed to the acceleration of cord cutting because one of the primary reasons viewers subscribe to cable and satellite is the opportunity to view live sports programming. The good news is we will have the Summer Olympics in 2021 as well as a more robust and predictable calendar of live sporting events that should engage sports fans and steer them back to broadcast television. Because we operate in primarily large markets in key swing states, we were able to garner a significant share of the total political spend across all TV stations. The total dollars here represent what was spent on our stations and at our television competitors, the share by station and how much of the total political dollars we garnered in each of our markets. As you can see from the left-hand column, our PowerHouse station in Detroit earned a whopping 36.4% of all TV political, and that is directly attributable to its incredibly strong TV news ratings and web dominance. All GMG stations performed well and over-indexed their traditional advertising shares when it comes to political. Combine that with competitive races in Florida, Texas and Michigan, and you can see why we wound up having such a strong year. Another way to look at our positioning is to examine our average revenue per station against some of our peer companies. While we may operate 7 stations in only 6 markets, we operate within larger markets and can do so very effectively and efficiently. Television stations have a relatively high fixed cost, but revenue increases exponentially as you consider market size. Many of our peers operate in much smaller markets, where the opportunity to earn revenue is limited by market size and dynamics. They need a larger number of television stations to realize a similar amount of revenue to what we capture in fewer but larger overall markets. Television news ratings and digital access across the GMG footprint remain strong and are a meaningful reflection of the critical role we played this year in informing, uplifting and celebrating our communities as we all struggle to grasp the enormity of the pandemic on every aspect of our lives. Local news has never been more relevant and remains a highly trusted and sought after source of information for a large portion of the news consuming public. Thank you very much for your time today, and I'm happy to answer any questions you might have during the Q&A session. And with that, I would love to turn it over to Jake Maas.

Thank you, Emily, and good afternoon, everyone. I'm Jake Maas, and I have been Senior Vice President of Planning and Development at Graham Holdings since 2015. Tim asked me to provide an overview of Graham Holdings' manufacturing segment. In addition to talking about the manufacturing segment in aggregate, I will also provide an overview of the businesses that make up this segment, Hoover Treated Wood Products, Dekko, Joyce and Forney. We think this is a great collection of businesses with deep moats and sustainable earning power. And I look forward to highlighting each respective business for you today. In Q2 of this year, Graham Holdings acquired Framebridge, which has a large manufacturing component to its business model. While Framebridge does not fall within our manufacturing segment for reporting purposes, given the manufacturing component to its business model, we thought investors would appreciate an overview of that business as well. I will conclude by providing an update on the impact we are seeing in each of these businesses as it relates to the COVID-19 pandemic. Not long ago by Graham Holdings standards, approximately 7 years ago, we had no manufacturing segment whatsoever. So I thought it might be useful to start from the beginning and explain our rationale for adding this segment. The logic was straightforward. We sought out and ultimately found businesses that we believed were durable business models that are easily understood; had track records of sustained earning power that would continue under our ownership; possess great leadership teams that were interested in continuing with the business; had attractive reinvestment opportunities that we believe would surface over time; and finally, were able to be acquired at a fair price that would create attractive cash-on-cash returns for our shareholders. We are proud of the fact that in a relatively short period, we have been able to find businesses that met these criteria. We would love to find more businesses to layer into the segment, whether as bolt-on opportunities to one of our existing manufacturer businesses or as new manufacturing platforms. That said, if we are unable to find good opportunities that meet our criteria, we are also perfectly happy to simply remain good stewards of our current businesses until the right opportunities surface. The manufacturing segment made up $411 million or 14% of the overall revenue generated by Graham Holdings in the last 12 months, ended September 30, 2020. And while this segment has certainly been adversely impacted in 2020 by the COVID-19 pandemic, a topic I'll address later in this presentation, we were pleased that our manufacturing businesses were able to generate $43 million or nearly 1/4 of our adjusted operating income for the last 12 months ended September 30. Our manufacturing segment has clearly grown to be an important part of our business at Graham Holdings, improving the stability of our company overall and strengthening our long-term prospects. As I mentioned, the manufacturing segment at Graham Holdings currently consists of 4 businesses. Hoover Treated Wood Products, Dekko, Joyce and Forney. In terms of relative size, you can see that Hoover made up 45% of the segment's revenue over the last 12 months; followed by Dekko at 40%, then Joyce at 10%; and finally, Forney at 5%. On average, the adjusted operating income margin profiles of these businesses, on a combined basis, are around 10%, a level that is understated by the fact that Hoover's revenue includes wood as a pass-through cost. We are pleased that our manufacturing businesses have been able to maintain their margin profile this year. And we are hopeful that as we scale these businesses, we will see upside to those margin levels once we get to the other side of the pandemic. With that high-level overview of the manufacturing segment, I'd like to highlight each respective business, starting with Hoover, a business we acquired in 2017. In business since 1955, Hoover supplies pressure impregnated kiln-dried lumber and plywood products for the fire retardant and preservative applications. Hoover operates 10 owned -- company-owned wood treatment facilities located across the country that service 100-plus member stocking distributor network that covers the U.S. and Canada. Last year, Hoover expanded its footprint by launching a new greenfield facility in Havana, Florida. The primary end markets for Hoover's fire retardant wood products, which comprise over 80% of Hoover's revenue, are with constructed multifamily homes and commercial buildings. Essentially, any wood structure built in the U.S., that is larger than a single-family home or has multiple tenants, requires the use of fire retardant wood by local regulations and building codes. Hoover has several aspects to its business that we believe make it extremely durable, including the fact that it is the definitive market leader in the space and the only player that truly has a national footprint capable of serving larger national customers. It is also the only vertically integrated player with scale that both produces its own chemicals and manufacturers wood products utilizing those chemicals. The result is that Hoover is the low-cost supplier in the space, with a brand that is nearly synonymous with its key product. Furthermore, Hoover's products make up a small portion of the overall cost of a project, but serve a mission-critical role in the overall safety of the structure being built. The combination of these factors make Hoover a great business with deep moats that is operating in an attractive and protective end market. Dekko, acquired by Graham Holdings in 2015, is a manufacturer of customized electrical equipment, including power and data solutions for installation in office workspaces, architectural lighting and electrical components and assemblies. Under our ownership, Dekko completed 2 meaningful bolt-on acquisitions, ECA in 2016 and Furnlite in 2018, both of which operate within Dekko's power and data segment. In 2019, Dekko launched a new lighting brand, LUX Iluminaire, which has gained meaningful traction and has the potential to be a growth driver for the company over the coming years. The primary end markets for Dekko's assortment of products include office furniture OEMs and dealers, hospitality, commercial real estate, white goods, transportation and durable medical equipment. Dekko focuses on manufacturing low volume, high mix, high complexity products that are not easily commoditized by non-U.S. manufacturers. In addition, Dekko's flexible manufacturing footprint, with operations in both the U.S. and Mexico, enabled the company to efficiently and cost-effectively meet the different needs of its customers. While the business has been more affected than our other manufacturing businesses by COVID-19 due to its exposure to commercial real estate and hospitality end markets, we believe the company is well positioned over the long run to continue to be a leader in its respective markets. We acquired Joyce in 2014, and the business has been in continuous operations since 1873. In 2019, Joyce acquired Uni-Lift, which had a complementary line -- product line of screw jacks. Joyce has now integrated those products into its existing manufacturing facility in Portland, Indiana. Joyce's main product offerings consist of screw jacks and linear actuators. To put it simply, Joyce manufactures products that help their customers lift, position and hold up to 250 tons for hundreds of applications around the globe. Joyce has a diversified set of end markets given the widespread utility of its products, but had somewhat heavier concentrations in food industry equipment, communication antennas and conveyor equipment. One fun fact about Joyce is that they are still providing the same jack to the railroad and mining industries as they did when they were founded in 1873. That is not to say that Joyce has not had to innovate to better serve their customers. A business cannot continuously operate for over a century without innovating. But it is also clear that the company benefits from a tried and true set of products that we believe will continue to stand the test of time. We see the key moat inherent within Joyce's business being the quality and reputation they have built up in their brand over many decades, the lower degree of technology risk associated with their core products and a U.S. manufacturing footprint that is difficult to disrupt by non-U.S. competitors given the nature and scale of the products they produce domestically. Forney was the first business we bought in the manufacturing sector in 2013. Forney has manufacturing operations in Mexico, and produces 4 main products: burners, igniters, dampers and controls, utilized in coal and natural gas power utilities. In addition, Forney provides aftermarket parts and services to its utility customers. For almost 100 years, Forney has provided safe and innovative combustion products and is helping the utilities to transition from coal to clean burning natural gas. Like Joyce, Forney's key moats included stellar reputation for high-quality products that go into one of the toughest, most demanding environment one can imagine, power plants. Going forward, Forney is focused on accelerating its sales efforts around its higher-margin products and services. The last business I'd like to highlight is the most recent addition to Graham Holdings, Framebridge. Again, Framebridge does not fall within our manufacturing segment, but manufacturing is a key component to its model, so we wanted to highlight the business today. Framebridge is different from other acquisitions we have done in recent years and is an example of Graham Holdings being willing to suppress short-term profits for long term gains. We did not acquire Framebridge because of its ability to generate profits in the short term, but rather out of the belief that 10 years from now, it could be the most popular way in the U.S. to custom frame. Framebridge's founder, Susan Tynan, founded the company with a simple goal. She thought the cost and hassle of framing objects at traditional custom framing shops was too expensive, outdated and a poor customer experience. She thought there was a better way, and Framebridge is the manifestation of that vision. Framebridge differentiates itself in 3 key ways. First, Framebridge has a unique operating model that leverages a centralized manufacturing strategy to gain economies of scale that allow them to have a lower cost model relative to their competitors that do their framing on-site at their retail locations. Second, Framebridge has an omnichannel strategy with strong e-commerce capabilities enabled by their scale. And third, because Framebridge does not need to catch a framing workshop to their retail locations, they are well positioned to create a more optimized retail footprint that meets their customers where they are, whether that is online or at well-trafficked retail locations. The result is that Framebridge has created a new kind of framing experience that is modern, consumer-friendly and less expensive, enabling them to not just take market share but to grow the overall market by bringing new customers into the segment. While there are certainly complexities in the business, understanding the model for success is quite simple. Framebridge has 2 fixed cost structures, corporate costs and manufacturing. It needs to sell enough units to cover these 2 cost centers. The higher the gross profit per unit, the lower the number of units that need to be sold to achieve breakeven and generate free cash flow. The company is not yet at the volume for breakeven, but has a clear path forward with understandable unit economics. The value proposition is clear as well. It offers lower prices, more convenience, faster turnaround and quality that is equal to or better than the competition. Thus, while Framebridge will require a significant amount of investment over the next few years before generating free cash flow, we believe the business is well positioned to scale and to become the category leader over the next decade. We are excited to be the new owners of Framebridge. Before concluding, I want to provide a snapshot overview of how COVID-19 has impacted our manufacturing segment in 2020. Like Graham Holdings, in general, our manufacturing segment is diversified with a wide variety of end markets. Thus, the impact of COVID-19 has varied at each of our businesses. The one thing that is consistent across all of our manufacturing businesses is that operating the manufacturing business in the middle of a pandemic is difficult. While remote work has become the new normal for many workers, the reality is that you cannot manufacture goods remotely. Thus, we are very proud of our leadership teams for putting the safety of our workers first, and overhauling their operations to allow their facilities to continue to operate throughout the pandemic. That said, our operations have not been disruption-free. And our teams remain vigilant as we continue to navigate a challenging environment. Make no mistake. We will do our very best to keep our facilities open and operational, but we will not compromise on worker safety in order to do so. In terms of our key end markets, the impact to our businesses have differed by company. For Joyce, Hoover and Forney, the impact of their end markets was relatively moderate. Combined, these 3 companies saw revenue decline only 1% for the first 9 months of the year relative to 2019. While this seems minimal, we believe our revenues would have grown in these businesses during the first 9 months as the pandemic now developed. As mentioned previously, Dekko has seen their end markets more severely disrupted, given their exposure to commercial real estate and hospitality. Dekko's top line saw a 24% decline in the first 9 months of the year. Framebridge, on the other hand, has seen COVID boost its online sales, which currently comprise the majority of its revenue, as remote workers across the country have decided to utilize Framebridge's custom framing services to spruce up the walls of their new home offices with memorabilia, diplomas and children's artwork, just to name a few of the myriad of items the company frames each day. Its retail locations, which, again, make up a minority of its revenue, are all open for business, but seeing reduced traffic as they mandate proper social distancing and other safety protocols within their locations. The good news on the retail front is that as the company looks for attractive new locations, the retail landscape has shifted in their favor. Excluding Framebridge, on a comparative basis, you can see that our manufacturing businesses are down year-over-year for the first 9 months of the year by 11% on the top line and 10% on the bottom line. At the start of the year, pre-COVID, we would have considered this to be a disappointing result. On the other hand, in the middle of the nationwide lockdowns that were occurring in the spring, we would have looked at these results and consider them to be a great outcome through the first 9 months of the year. From where we sit today, we are pleased these businesses have so far proven to be a durable earnings center for Graham Holdings, even through a very difficult macroeconomic environment. In conclusion, the manufacturing segment is a solid new leg of the stool in terms of earnings diversification for Graham Holdings. While this segment has been adversely impacted by COVID-19, we believe the earning power of these businesses remain strong and durable over the long term. Over the last 5 years, we found good reinvestment opportunities within this segment, both in the form of bolt-on opportunities and organic investments and expanded capacity and new products and services, and we anticipate being able to find additional attractive reinvestment opportunities within this segment over the coming years. We're excited to have added Framebridge to Graham Holdings as we believe it will prove to be a great long-term investment for our shareholders. Finally, we will continue to opportunistically look to add new companies to this segment to the extent we can find businesses that meet our acquisition criteria. I'll now turn it back over to Tim.

Thanks, Andy, Emily and Jake. We're fortunate to not just have these great businesses, but to have the best-in-class running them as well. We hope that these presentations have provided better insight into our operations, our thinking and our path forward. We'll now move to the Q&A portion of the event. There's a chat that people can enter in questions that they may have, and some have along the way. And in addition to the presentation team, with us today are Wally Cooney, our CFO; and Nicole Maddrey, our General Counsel. We received several questions in advance, and so the management team and I will respond to those first. And once again, if you'd like to ask a question, please use the chat feature on the screen to type it in. And we'll take as many questions as time will allow.

And so the first question, the other businesses division produced losses of $20 million to $30 million annually during 2017 to 2019 based on EBITDA. We assume that the auto dealership and Clyde's were EBITDA positive in 2019. Slate and foreign policy have been around for a while, so we assume any losses there would be modest if there were any losses at all. That leaves Megaphone, Pinna and CyberVista. It would be great to get some color on the losses from these companies. We trust management to make prudent decisions. On the other hand, these are large annual investments, given Graham's current market cap. What is the reasonable expectation for losses from Pinna and CyberVista going forward? So thanks for the question. I would say that I generally agree with several of the characterizations, and I can try and provide a little bit more color. So I think the -- correct that, generally, the automotive dealerships and Clyde's are profitable, non-peak COVID times aside. A couple of other things to call out on this front. Megaphone had losses, as discussed earlier in the presentation. Assuming that, that transaction closes, the -- those losses would be removed on an ongoing basis. So people can factor that in. On -- 2 other comments on this on Pinna and CyberVista specifically, we haven't broken out exactly what those numbers are. But the thing I would say is my expectation is that losses for those businesses have likely peaked and that you would see a reduction in investment level on a go-forward basis. And then the only other thing to consider when looking at this other businesses segment is we did acquire Framebridge midyear in 2020. So -- and as we've referenced several times, we view that as an investment stage business really where the data tells us we should be investing in it pretty hard. So that will be part of that segment on a go-forward basis. So hopefully, that's a little bit of color there that that's useful for people. The next question, I'm going to ask Wally to answer this, which is have the rating agencies communicated what Graham would have to do in order to receive an investment-grade rating. Wally?

Okay. So Moody's and S&P update their reviews of the company, at least on an annual basis, sometimes more frequently. In terms of an investment-grade rating, to summarize, I would say this -- the company has credit strengths that are important factors, a strong balance sheet and strong liquidity, cash and marketable securities balances that exceed outstanding debt with a largely unused revolver and a significantly overfunded pension plan. Some of the areas the rating agencies would like to see in order to consider an investment-grade rating includes increasing the overall size of the operations of the company with stable and growing revenues and earnings power over time with the newer businesses and building greater scale in certain of the company's sectors.

Okay. Thanks, Wally. The next question is around the manufacturing segment. During the annual meeting earlier this year, management stated that our manufacturing segment is very well positioned to emerge stronger when the COVID crisis passes. Why does management believe this? More specifically, it seems that both Dekko and Hoover are exposed to commercial real estate, where COVID has potentially created some longer-term structural challenges. Will this continue to be a headwind for the manufacturing segment? So I think I might ask Jake to chime in on this one, too. But I think we think there's a couple of things. One, there certainly is some exposure to commercial real estate. It tends to be one of the markets our businesses are exposed to, not the only market, with a bit more at Dekko. That said, broadly, we tend to be less the type of real estate that is built downtown. So depending on how trends play out, that can factor in. But we're -- I think we're still in a watch and learn phase. The other thing that I think is -- maybe gets to the core of the question a bit more is the management at Hoover has been there for quite some time, decades, and they've seen a lot of ups and downs. And their experience is whenever there's a downturn, Hoover tends to pick up a little bit of market share. And so we're hopeful that in a downturn environment, again, the same trends would be true, and we could potentially emerge in the market in a stronger position than we entered. So Jake, anything else you'd add on that?

No. I think you covered it well, Tim.

Thanks. Okay. As always, we get -- we'd love to get management's updated thoughts and outlook for the broadcasting stations. Does the recent sale of ION have any relevance for Graham? It looks like a different business model to us. Did Graham look at the divestiture of the stations by Scripps? We understand that Graham stations are well-run, produce a lot of cash flow and have a low cost basis, but any updated thoughts on whether Graham should perhaps participate in the consolidation wave in the industry as a buyer or as a seller? That would be great. There are a few questions that came in, in the same vein. So I'll -- if your specific version was not asked, we'll cover it here. So on the first part of it on the ION piece, kind of agree with the questioner here. It's a pretty different model. And so I don't think it's really directly relevant to Graham Media Group. On our television stations and Graham Media Group broadly, I mean obviously we continue to stay very close to the trends of our business and the trends of the industry. As we've said before, we're very comfortable having a well-run business that produces a lot of cash for the business of for our shareholders and we look at opportunities as they come up and stay very close but we have been and continue to be, at this point in time, pretty comfortable continuing to hold the business. So obviously, we'll let people know if that changes, but a version of the same answer that I've given when this question has been asked in the past here. The next question. Is management willing to give any additional details on the company's marketable equity securities portfolio, either the composition or more color on the longer-term strategy? For instance, is management looking to add to this portfolio opportunistically? Or should we think of this as low basis positions that management plans to simply hold until better acquisition opportunities come along? So there's -- it seems like there were 2 parts of this question. One is on are there additional details? Something that we're looking at, potentially more to come on that in the future, but just something that we're looking at right now. In terms of how we view this, we view these -- less as a collection of a portfolio and more of as investments in individual businesses that we plan on holding for a long time. And we very much view them as buying pieces of businesses, and we tend to have a relatively long hold period. We have very much have a preference for wholly owned businesses than fractions of businesses. So to the specific part of the question, would we hold until a better acquisition came along? If we thought something where we had -- was really a great acquisition for the business and we had to sell down some of the portfolio to help fund that, we'd absolutely look at doing that. All right. The next question is about health care. Is the health care division competitively advantaged? If so, how, what is the long-term vision for this segment? How large can it scale? So taking the last piece of it first, it's a very large space. Obviously, health care is a large space. Our particular segment that we operate in is a pretty large space. So we certainly think it can get bigger. How large, it's really hard to say. I would say we are -- our business is -- we are mid-sized nationally, but we are very scaled regionally. And in this business, they are really regional businesses. And so where we operate, we tend to have a very large market presence. We also think that the world is moving in the direction of in-home care. So from a -- and overall, are the trends with us in this space? We think that they are. We think it's also pretty important and a bit of a competitive advantage to have good relationships with hospital systems, whether that's in the JV type of partnerships that we referenced earlier in the presentation or in other ways. We think that, that's a pretty important part of the ecosystem. And actually frankly we think it's a bit of a competitive advantage to be owned by a company like Graham in terms of whether that's being able to partner with hospital systems, or really think about what are the right investments that need to be made to make sure that we're having good business results, but good results for the care we provide as well. And that over the long term, those things go pretty hand-in-hand. So we think that we're a very good owner for this business. It's in a large and growing space, and we are scaled in the geographies that we operate, and we're hopeful we can keep growing over time there. Okay. The next question is about SPAC. We've been watching these SPAC craze and recently began to wonder if Graham might kick around some kind of SPAC offering? Short answer here. No. We're a very small team. And I don't know why we would be competitively advantaged relative to the hundreds, if not thousands of SPACs that may come over the coming years. So it's not in our -- not written down on a sheet of paper anywhere. Okay. We think that Graham remains undervalued, your thoughts? Well, we never really speculate on our stock price, and so I'm not going to start here. The only thing I'd reiterate is we did -- earlier in the presentation, we talked about repurchases and what our philosophy was associated with that. And just to reiterate, over the course of 2020 we really repurchased a substantial amount of stock. And so we felt that those criteria had been met. So it's probably the extent of the commentary around valuation and stock price that I want to give. But hopefully that's helpful. Okay. Wally, maybe you can take this next question. Please discuss the December 2019 transaction in which Graham Holdings purchased an irrevocable group annuity contract from an insurance company for $216.8 million to settle $212.1 million of the outstanding defined benefit pension obligation, whether you anticipate similar transactions in the future and how you think about the value of the pension overfunding in relation to the intrinsic value of Graham Holdings? Wally?

Okay. So the group annuity contract transaction that we closed in December 2019 was an additional derisking initiative for the company's pension plan and it effectively settled pension obligations covering certain retirees. It also had the added benefit of reducing PBGC premiums and administrative fees for the pension plan on an ongoing basis. And we've done other things in the derisking area in the past, and we'll continue to monitor and look at opportunities to further derisk the pension plan in the future. Regarding how we think about the value of the pension plan, Tim talked about this earlier, mentioned several ways that the company benefits from the overfunded pension plan in terms of annual cash flow savings, for employee benefits, funding employee reduction programs and the potential use in future transactions. So pension plans are designed to benefit current and former employees. And other potential uses of assets beyond that could be subject to excise taxes of 50% as well as income taxes. So that's certainly taken into account in how we think about the pension plan.

Okay. Thanks, Wally. Next question is a little related to an earlier one. But our stock is down about 20% over the past 3 years, while earnings are up nicely. Why are we lagging the market? And what steps are being taken to enhance stock price? I think it's hard to speculate on stock price at any given point in time. So I would say 2 things here come to mind. One is look, over the long term, we think if income goes up and share count goes down, that's going to be pretty good for long-term shareholders. And the second thing, I would just reiterate part of my answer on the repurchases, you can sort of get a sense of -- at the time of repurchasing earlier this year that we thought it was a good value for shareholders. And so we did that through Q3 in quantities. So hopefully that added some helpful context about how we view that. Okay. The next question, what's going on with SocialCode and why did you break it into 2 companies? So good question. SocialCode, just as a reminder for folks, is a company that we started about 10 years ago at this point, and really initially started by helping large brands help with their digital advertising and social platforms. And it's expanded beyond social platforms to marketplaces like Amazon and creative services. And over the course of this year, it became clear that SocialCode has been investing in a SaaS-based software product, which is now Decile. And it became clear that there was an opportunity with Decile and where we may have thought there might have been a lot of similar customer base overlap that, that was proving to be less true than we thought when that initially was launched. And so it kind of made sense for the business to start to focus a bit more on the things and the customers that they were good at. So we decided to split the 2 parts of the business into their own separate operations, so they can have a little bit more of that dedicated focus. So as part of that as SocialCode had moved beyond being just a social business, the name was something that increasingly was probably boxing them in relative to what they could do. And so SocialCode had a rebranding effort and changed the name to Code3 just a couple of months ago here as well. So that's a little bit of kind of what's going on there. So we now have the 2 pieces that had made up SocialCode. One, now Code3, kind of the traditional business around helping brands market on digital platforms. And then the second, Decile, which is a software-based business that really is the -- helping companies bridge the data gap between marketing and finance and trying to be the single source of truth for organizations when they're making both financial and marketing decisions. So that's a little bit of what's going on at SocialCode. Okay. Next question coming in is about Framebridge. Could you be more specific about your growth plans at Framebridge? Are you pursuing more physical store openings, such as you were doing in Brooklyn and Atlanta? Or is the greater opportunity online? So we think that omnichannel is really the right answer for this business, that there is the ability to meet customers in both fashions that the kind of traditional frame shop experience is very antiquated. And there are also a segment of people who are very comfortable doing it online, particularly if they're framing digital items. So as Jake walked through COVID, definitely accelerated some of the online piece of things and slowed down some of the retail piece. But we expect over time they both will be very big part of the equation. And where we have launched retail locations, we currently have 5 open, which we see an acceleration in the online business in that area as well. And so those retail locations serve a little bit as billboards. So in the coming year, we will continue to expand the retail footprint as well. Our expectation is that we'll likely open roughly 10 additional stores in 2021 in both existing markets and then also in a handful of new markets. So we will be going from 5 to roughly 15 by the end of next year. And we think the overall opportunity in the market is not capped at that number, that there's quite a bit of runway beyond that to go. So hopefully, that's a little bit more specific and helpful context on the growth plans at Framebridge. Okay. The next question is for Emily. How has cord cutting impacted our TV markets during 2020? And on how to retrans and reverse comp agreements compared to our public peers? On a year-to-date basis, what percentage of Graham Media Group revenue is digital? So Emily for what we can answer on that, why don't you take it from here?

Okay. Thanks, Tim. Well, first of all, with respect to cord cutting, as I said in Slide 10, we've seen traditional MVPDs go down. June over June, we showed a 12.5% decline but I would qualify that by simply saying that a significant piece of that is related to satellite carriers. So they disproportionately account for much of that cord cutting. And then that's mitigated, of course, by the increase on the virtual side, the virtual MVPDs, which are up 50% over the same period. Obviously, different scales, but there is some mitigation there. So it's not terribly different in 2020 or doesn't appear to be so far from other years, but it is something we've seen grow a little bit over the last couple of years. Again, mitigated by the virtual MVPDs. So that's good for us. On the retrans and reverse comp agreements compared to our peers, we don't have obviously specific numbers, and we wouldn't be able to share those. But we believe from what we are able to glean that we are quite competitive because we are in larger markets and have important affiliates with respect to NBC, for example, in Houston and Detroit, and the same goes for Orlando and San Antonio with CBS and ABC. So we have what we believe to be very fair deals relative to what our peers do. And the same would go for the retrans deals as much as we are able to ascertain. And then in terms of GMG revenue, what percent of it is digital, I would only speak to the percentage of our ad revenue, which is how we would measure digital. And it's roughly in the 9% category.

Great. Thanks, Emily. Actually, the next question for you -- is for you as well. And it's a follow-up on the digital piece, which is asking about the revenue side. But also is the revenue -- is the digital revenue source profitable on its own?

Yes. So thank you. In fact, our digital operations would be the equivalent of one of our midsize television markets in terms of profitability. So it is a very successful entity in its own right.

Great. Thank you. Okay. Next question, a quick one. When will we make these slides available online? They will be available shortly after this wraps up and certainly by close of business. So you should have them shortly if you'd like. All right. Our next question is how would we recommend investors value the other businesses segment? I would say it's not in my business to tell investors how to do valuation. So I don't have a -- different people do things very different ways. So I don't really have a good advice for somebody on what people view very, very, very differently, so not something I feel like I can provide a real productive answer on. Next question is, Event is very well put together. Thank you. My question is thus, why do you not have a quarterly conference call with investors like nearly every other public U.S. company? So we have not had quarterly conference calls for the existence of the company, and we also have not issued guidance. And we've generally thought that those things have served pretty well. And that most of our shareholders are comfortable with that fact, knowing that we manage the business for the long term. And if you care about, and this is something Don has said many times and I agree with quite a bit, is if you care about any individual quarter's results in an outsized way, then you probably shouldn't own the stock. So we think it's important to be able to provide updates to our shareholders. With our type of business and the pace at which things change and with our diversity, we've tended to think that a couple of times a year is sufficient. You'll note that this event is almost exactly 6 months from our annual meeting, which we have turned into an increasingly robust presentation of what's happening with the business. And so we're comfortable and have gotten feedback from a lot of shareholders that they're comfortable with the level of presentation that we've been providing. Of course, people can always call after any quarterly release if they have any clarifying questions, but that hopefully provides a little bit of color on what we do and why. Okay. The next question is do you believe Kaplan can achieve double-digit operating margins? Does that need to occur in order to validate it as a good business? Andy, I might have you chime in here as well. But I do believe that double-digit operating margins in Kaplan are possible, I think, as the business has moved more digital. You do have the characteristics that mean future revenue growth likely is a bit more accretive. But I think you could have a business with less than 10% operating margins that can still be a very good business. The higher you have, the more cushion you certainly have and the more degrees of freedom to invest and still generate cash flow, however, when it's necessary. So Andy, anything you'd like to add on that or any extra color?

I certainly agree that it can get into double digits. I would just note that like the rest of Graham Holdings, we manage the business for the long term. And that means making sure that student outcomes and student quality predominate. And we will always keep that focus to sustain the business for the long term. So -- but sure, we can get to double digits, I believe.

All right. Next question is what do you plan to do with the proceeds from the Megaphone sale? So we don't have the proceeds yet, so we try not to spend it before we have it. But in general we'll continue to operate like we always have, where if we see something that we think makes sense where we don't hesitate to act. But we're fine being conservative and we're fine being patient. So there's no specific set plan associated with those proceeds, other than we'll continue to operate like we always have. Next question. So how do you have enough management bandwidth to track, manage and allocate capital to such a diverse array of businesses, each with their own set of issues? Is it enough to simply pick great people to run them? I like the question a lot. This is something we think a lot about internally. And the questioner, I think, answered it a little bit in the end, which is when they said, "Is it simply enough to pick great people to run them?" That helps an awful lot when there are operators of the business that truly understand the industries they operate in. And therefore it doesn't require me and the team here to get engaged on a deep operational level on a day in, day out basis. It really does free up capacity to think about the future and to think about where our opportunities and where to put capital. So that is a big piece of things. I certainly think that there's a limit to what we do. I don't think we've hit that yet, but I certainly think there's a limit. Occasionally, you get things that come off the plate. Megaphone was an area that I would say I personally spent a good amount of time on, and so did many, many other people in the corporate team. And that is a piece of things that won't be part of the future after deal close. So we do think there's a limit. We don't think that we have reached it. And the biggest way that we are able to kind of get leverage is by having great people to run the businesses. Okay. The next question is a quick one. How big is the custom frame market? So the custom frame market, I'd say there's 2 things to think about on it: what is the custom frame market today, and what do we think it will be in the future. Today, we think it is in the 3 billion to 4 billion annual sales range domestically, and that it's broken down with some percentage of sales at a couple of big-box retailers and then a lot of small individual stores. We do think that with a model like Framebridge, the market will actually grow, and that as Jake referenced in his presentation, we'll be able to go. And we believe we can take market share from that existing market, but we think due to price and due to frankly convenience and service and turnaround time, that we'll actually bring more people into the market. So that's a little bit of a view on kind of what it is today, but we actually think with our model, it can grow not insubstantially as well. Okay. The next question is for Andy. To what extent do you see the growth in Purdue Global enrollments is a short-term increase from COVID-19 versus a longer-term trend growth trend for the business?

Yes. Well, I think I'd have to break the impact of COVID into 2 factors. One is the economic anxiety associated with concerns about whether people are able to keep their jobs. And that is a -- you all can decide how long people will have that economic anxiety, but it's not forever, I certainly hope. On the other hand, the openness to online which COVID has enabled, suddenly every learner in the world became a digital learner. Every faculty member in the world became a digital instructor. That has opened people's eyes to the opportunity and the potential quality of online education. And that's something that I think is going to be sustained and going to accelerate over the long term. And that is our job to make sure that Purdue Global and other institutions that we support get their fair share, or more than their fair share, of that increased goal and interest. But I think that the net of COVID is a long-term plus.

Okay. Thanks, Andy. Emily, the next question is for you here. What are you doing differently than your TV station peers that has allowed you to have industry-leading TV station EBITDA margins for the last 20-plus years?

Well, thanks for that question. So I would say we operate in a really interesting sweet spot because we have relatively large-sized markets with very strong news operations. And we are not so big that we need a lot of corporate overhead, so we have a very lean corporate staff. And we have had, for many many years predating me and going way back, a culture of innovation and creativity. And we are relentlessly focused on serving our communities. So I think when you put all that together, that has been a very large piece of why we've been able to operate so efficiently and effectively in all of our markets.

Okay. The next question is how do you prioritize the financial goals for the holding company? Are you attempting to maximize book value per share growth? Or do you believe growing free cash flow per share -- Or do you believe in growing free cash flow per share or some other metric? Based on whatever metrics you believe make the most sense, how would you grade the last 3 to 5 years? So I think ultimately we are looking to grow intrinsic value per share. And I think over time, that is going to be measured on a free cash flow basis combined with what other assets do we have, whether it's marketable securities, et cetera. So I think we are willing to as we've talked about, we're willing to depress on a short-term basis if we think that the long-term cumulative free cash flow that we generate will be higher by doing so. But we really think that it's an intrinsic value per share and growing free cash flow per share will be a pretty good -- a measure of that over an extended period of time, combined with the other assets that we have that may be nonoperating assets. Okay. The next question is for Andy. What percentage of remaining Kaplan Test Prep revenue is digital versus retail?

Well, I think this question is really digital versus face to face. I think that's what you're asking. And we're now 100% of our test prep is digital. And coming out of COVID when face-to-face locations reopened, we expect the vast majority to remain digital. We are closing centers. And really students, they are much more accepting of online. They seem to be happy with it. It's got a lot of convenience benefits, cost benefits. And so I would expect the great majority of our test prep to be digital going forward.

Okay. Next question is will Framebridge be a franchise model? The Framebridge is not a franchise model. And I -- it's hard to predict the future, but certainly for the foreseeable future, there are no plans that it would be. We think there's a lot of value in kind of owning the operational experience right now. So hard to say long, long, long term, but for the foreseeable future there's no plan to franchise. Okay. All right. I think this looks like it might be our final question. Jeff Bezos said he got a great shareholder question. "I have 100 shares of Amazon. What do I own?" I would like to hear your answer to the question. I own more than 100 shares of Graham Holdings. What do I own? I like that question very much. So when you own Graham Holdings, you own a diversified business that continually tries to increase its durability, increase its competitive moat and its leadership positions in our chosen fields. And that business that you own, it's run by owner operators who really look to grow intrinsic value per share at adequate rates over time. You own a business that looks to serve its customers and its communities and its students with honor and integrity. Every day, we work hard to operate a great business, but also to make a meaningful contribution to society with the services we provide and the products that we sell and that we make. For longtime shareholders, it will come as no surprise that the company has been run this way for decades, and we aim to do so for decades to come. So hopefully that gives you a little bit more sense of how we think about what you own. So with that, I'd like to thank everybody for attending and really thank everybody for the very thoughtful questions that people submitted in advance and also during the event. And once again, thank you for attending today's event.