|Shares Out. (in M):||95||P/E||7.6||8.15|
|Market Cap (in $M):||3,684||P/FCF||7.6||8.15|
|Net Debt (in $M):||3,170||EBIT||345||276|
Sign up for free guest access to view investment idea with a 45 days delay.
I believe that an investment in TRCO offers the opportunity to profit off of the longer tail in local, rising net retransmission fees, several hidden assets and the downside protection of repurposing the broadcasting assets should the national advertising secular decline intensify. I believe the core business is worth $55 with additional upside from the real estate (+$9.76) and spectrum portfolio (+$7.02), or 85% total upside, as shown below. I also think it’s particularly timely to highlight the value given a potential breakup of the company in a sale.
Current Characteristics of the Changing Media Landscape
Technology Change Results in Greater Dispersion in Value of Content
Given the commoditization of distribution mechanisms, the value of quality content is increasing while the value of non-exclusive syndicated content declines. This destroys value in commodity networks like the CW as viewers can watch the same content through OTT while it increases the value of networks with original programming – they have much greater negotiating power across the board since television still offers the best scale for national video advertising campaigns (i.e. creating a lifestyle brand).
Ad-Skipping Makes OTT A More Powerful Alternative
It’s clear that consumer preferences are changing and ad-skipping has had a bigger historical effect on the industry than OTT but I think that’s changing. As consumers shift to more ad skipping technologies, the content companies react by raising affiliate/retransmission fees to offset the difference and the cable companies are raising prices of bundled video offerings but the fundamental question is really how elastic this market is going forward. Can the cable companies keep raising prices to offset affiliate/retransmission fee increases or will cord cutting and the shift to OTT accelerate and is there a point where broadcaster revenue inflects downwards. That’s really the fundamental question that drives earnings power here, very big picture. SVOD changes are staggering, usage is up to 95min per day (Netflix usage data) and streaming subs are up 20%. That said, while local advertising revenue will not decline substantially, the growth of retrans/affiliate fees certainly can.
Not All Advertising is Equal
Advertising dollars are shifting around as the migration to digital occurs but some segments have had longer tails than others and the other fundamental question is if those tails fall off faster in the future than they have in the past. While national television advertising is in secular decline, local, political and sports have all held up and it seems like the tail is fairly long. Despite slightly lower viewership in local news, there’s no other medium by which local restaurants, auto dealers, retailers…etc can host a video based campaign with scale/reach. Political also seems to still see the most efficiency from local news advertising, especially in swing states. Sports has obviously held up due to its live viewing nature.
With Whom Does the Negotiating Power Lie, the MVPDs, Networks or Local Broadcasters
These relationships have shifted over time as technology change has caused quality content to become more important, as discussed above. The recent consolidation should result in better negotiating power for the MVPDs - taking channels off the air is more impactful when your footprint is larger. (For example, CableOne makes no margin on their video subscribers vs. 30%+ margins at the larger MVPDs.) The networks who own quality original content and national sports rights (CBS in AFC markets and FOX in NFC markets in particular) have better bargaining power than the lower quality networks (like CW). TRCO/local broadcasters have some negotiating power over the MVPDs since (especially with Aereo gone), the MVPDs need the local channels. They also have some negotiating power with the networks given the local salesforce and local news infrastructure in each market. Moreover, the networks cannot recreate the local model given the 40% station ownership caps. The local broadcasters also have the ability to switch networks, as TRCO did in Indianapolis which gives them some leverage over the networks.
Local Advertising Spend
SBGI has talked about how local television is only 15% of local spend while direct mail, print, radio and outdoor make up the rest. Direct mail is 28% of local and television has been taking share from it. Local broadcasting will continue to be insulated as it is the last of these local segments to lose share to digital because it’s the only place where you can create an overall campaign with a lot of scale. SBGI, MEG and GTN have also cited Honda as an example; they shifted most of their spend to Youtube in early 2014 but are in the process of shifting back because the returns were bad – online video viewership is at 15min/day vs. 312min for linear television and the age demographics are likely very different thus you will still need linear television for reach – while broadcast network audiences have declined at a 2.2% CAGR since 1990, broadcast advertising revenue is up 3%. From the NAB, we know that communications (mainly wireless) and auto are the biggest drivers of local ad dollars. Auto and wireless were both weak in 2014 but seem like it’s more cyclical than secular – and given new types of cars (aluminum body trucks) and new phones, video advertising will always increase sales.
I model -3%/-5% in odd/even year step downs (-4% CAGR) in national advertising between 2015 and 2019 given the secular arguments discussed above. I assume local is slightly healthier at an 8%/-9% (-1% CAGR).
The market is very uncertain as to how net retrans changes will affect the local broadcasters’ FCF over the next several years. With retrans only introduced in the last few years, the market initially heralded this high margin new revenue stream for the local broadcasters since they are 36% of the audience share but only 9% of the affiliate fee base. However, the networks began pushing for a percentage of retrans in reverse comp. A deal in Indianapolis where CBS demanded $0.80-0.90 in reverse comp caused Lin to stall negotiations and TRCO stepped in to take the affiliation causing significant stir in the industry. Same with Seattle where TRCO almost lost its Fox station but at the last minute, agreed to a reverse comp number estimated to be in the $0.70 range. In these scenarios, net retrans drops to 30% which we sensitize below for our FCF estimates. GTN believes they should do about 65% in net retrans over the next several years; TRCO has guided to 50% post their 2018 Fox negotiation while CBS is targeting extracting 50% by 2016 and 60% by 2020 (implying 40% net retrans for the local broadcasters). However, as per the discussion above around negotiating leverage, TRCO has much higher leverage with CW than the larger networks. The local salesforce, local news infrastructure and 40% cap (so the networks can’t just start their own local stations) all help to create barriers/negotiating strength so that it’s unlikely the networks take more than 70% of the retrans stream in the longer run.
I model this based on a breakdown per station. I then look at which MVPDs are not paying TRCO and normalize each station’s retrans per sub to get a value per paying sub then assume the subs not paying begin paying (though I model a 50% step up, assuming Comcast and Dish are paying some nominal retrans). From here I assume the next 50% of their base is negotiated between 2015 and 2017 and then this fully negotiated base steps up by 2% in 2018/2019.
We know TRCO paid $273m in broadcast rights fees in 2012. While we don’t know what percentage of this is reverse comp vs. syndication/sports payments, I assume about 25% is reverse comp as this gets me to a programming cost ex this reverse comp in-line with peers. Management has said they have a 10 year old contract with the CW and they are paying them more than you’d expect (more on that below). I think this will come down in 2016 since the CW has limited leverage over them (SBGI pays CW no reverse comp). This will help offset the increased LTV Fox fees in 2018. LTV said they were getting a 20% net retrans margin previously which comes to $0.20/sub/month. Assuming this steps up $0.65 to $0.85 (higher than Seattle example discussed above), we increase their payments by the total above.
The above gets me to about 52% net retrans margins by 2019 which is above management’s estimates but that’s where I get to based based the math – obviously they are paying less for CW/My. I also include the Fox stepdown in 2018 instead of 2019 because I want to look at normalized FCF for election/non-election years and believe 2020 is too far out.
CBS’s new OTT service operates only in the O&O markets but seems to offer the ability to use digital to create an over the top offering like Aereo. ATSC 3.0 is going to be very helpful for doing this as it allows for better signal propagation to mobile devices. CBS OTT is not a threat to the local broadcasters because CBS must get approval to stream in their geographies. I model digital increasing from 3% to 6% in several years as per management’s guidance. I believe this is possible given the OTT potential highlighted by SBGI.
The CW stations offer the most potential to shift utilization of the asset in the future if advertising declines – as per the chart below, with syndication at 52% of revenue, the CW stations would lose the most as viewership shifts to OTT – i.e. this is a terrible revenue stream and makes up 40% of TV revenue. The only way to be comfortable with an investment in TRCO is to believe this asset can be repurposed for downside protection. TRCO believes the CW stations have some value over independent channels but it’s limited. TRCO has some scale now with 30% of the CW network’s total distribution and their deal with the CW is 10 years old and will be renegotiated in 2016 at which point they can substantially reduce the fees they pay. As per the negotiating leverage section above, situations like Indianapolis offer TRCO the opportunity to put the CW on a multicast station or channel share though they would lose their satellite subscribers by doing the former.
I view WGN most as a cheap option where I assume continued higher content costs but that any major upside on affiliate fees above $0.095 by 2018 are simply a free option. While management turned around FX and was responsible for Nip/Tuck and Rescue Me, the technological change discussed above around increased distribution networks increase the competition for high quality original content and reduce the probability that a show succeeds. Between NFLX/Amazon Prime original programming, the pay TV channels, networks, cable nets…etc all developing original programming, the viewer has so many great options that creating an amazing show becomes a lower probability event. However, Manhattan and Salem have had some success. It’s certainly not impossible but much harder to do today. That said, converting to a cable net and eliminating non-exclusive syndicated programming will allow TRCO to capture the copyright fees in the $50m range that the MVPDs are paying through the government (of which TRCO gets a small percentage). I assume they are able to capture 100% of this through higher affiliate fees ($0.095 by 2018) and are able to grow distribution to management targets of 90m given their scale.
I assume original content costs per management guidance and 5 shows/year.
Very briefly, while they seem to be excited about their model here where they incur some production costs for shows but participate in off-air revenue streams and have more control over airing their own content across networks, as per above, everyone is getting into original programming so the odds of a hit go down. I don’t ascribe any incremental value here.
Liguori said this division was referred to as a hidden gem while he initially diligenced the company. Recurring revenues are high and they are making acquisition to increase scale internationally and in the US. That said, if Rovi is 5x their size and renewal rates are 90%+ and contract lengths 2-6 years, I don’t see how they can grow substantially in the US and if they are growing internationally, they will need to increase their foreign language content, thus are unlikely to achieve Rovi’s 85%/35% gross/EBITDA margins. However, the hidden value in TMS, I believe, is their ability to capture what viewers watch/listen to and combine and use that customer data when creating shows for Tribune Studios (another free option that’s hard to value). That said, I have not spent enough time on the TMS business and believe this is a good management team and given the recent acquisitions they’ve made in the division, they could drive value over time.
Cable/satellite channel 31.3% owned by TRCO, fully distributed in the US. Estimates are for flattish affiliate fees of $0.20 (up 3%) going forward so I assume TRCO cannot grow income from Food Network, especially because television viewership may decline as per the discussion above. I think the franchise value here is solid and continued payments are predictable. There were articles in the past criticizing Food Network for not paying their famous hosts enough but despite some of that controversy 8 years ago, the ratings have been stable. With flat viewership, slightly rising affiliate fees of 3-4% and likely flat advertising, Food Network should grow EBITDA at 3-4%. This is about $140m of the $200m annual distribution.
This is an online job board of which TRCO owns 32.1%. While LNKD is taking share from other sites, Career Builder is somewhat supported by the newspapers who own it as they link their own classified job ads to the site, giving it more staying power than a MWW. This is about $60m of the $200m annual distribution which I believe stays flat.
Assume $75m in maintenance capex as per management’s commentary. This puts TRCO at about the same capex as % of revenue as the low end of its competitors in broadcast while real estate should not change vs. the past. I model 40% as per management estimates. Interest expense should stay constant as management is happy with their debt level and the term loan is due in 2020.
Free Cash Flow
While the exact setup of the auction is unclear, TRCO has some options to unlock value from its spectrum. These include (1) participating in the auction and closing down certain underperforming stations, (2) channel sharing with (a) itself in duopoly markets or (b) competitors in single station markets, (3) converting UHF stations to VHF and receiving some proceeds, (4) holding onto its spectrum and setting up a broadcast LTE leasing stream to the carriers in the future. In evaluating (1), I look at the sensitivity vs. declines in FCF from closing down stations in order to understand what maximizes value but we can look at (2)/(3) as offering hidden asset value that does not impair earnings power. (4) relies on future technology changes that would allow TRCO to operate broadcast LTE on the same channel as their HD streams, thus is the least precise.
Before delving into valuing each option, see below for my framework in valuing TRCO’s spectrum by market.
Bold denotes markets where the NAB believes there is no value to the spectrum. These markets are not incorporated into any of my spectrum value numbers.
I believe the new FCC/Greenhill values likely overvalue the spectrum in order to entice broadcasters to participate. Since they are opening bids and it’s a reverse auction structure, it’s unlikely that the carriers will pay such high prices as $7-40 per mhz-pop when the most recent AWS-3 auction closed at an average bid of under $3mhz-pop. Instead, the clock will likely tick down to a similar price (adjusted for net proceeds, properties of the frequency and interference). As such, for all non-interference markets, I average the recent AWS-3 auction price in each market for 10x10 blocks with my estimate of spectrum values based on the cost to add the same amount of capacity by densifying the tower network in that DMA. (ie, if adding 6mhz of spectrum allows a carrier to add X new bytes in capacity, how many new towers would the carrier need to add with its existing spectrum base to add X new bytes to its network). I then assume the net proceeds to the broadcaster is 75% of this value (given the costs needed to fund FirstNet, broadcaster repacking equipment costs and pay broadcasters for UHF-VHF conversions/interference market differences. See appendix for more detail.
(1) Closing Down Underperforming Channels to Sell the Spectrum
While I hoped this time consuming analysis might yield a discovery that there are several underperforming stations that should be shut down and sold for spectrum value, the results show the obvious case that aside from two DMAs (Milwaukee and Oklahoma City) it makes sense only to close down interference markets where the FCC will likely reward broadcasters with higher proceeds than the per mhz-pop bids in that DMA due to interference in nearby DMAs with much larger populations. Assuming our interference assumptions are correct, though they need to be much better vetted, closing down just 8 markets yields significant proceeds from the auction with little OCF loss, a 16x multiple on OCF.
(2) Channel Sharing
Technological changes in broadcasting around modulation have improved equipment efficiency, allowing stations that previously required a 6MHz channel to broadcast HD/SD transmissions to now broadcast in 3Mhz, ie multiple networks can share a channel. In early 2014 in LA, KLCS and KJLA successfully tested transmitting two 720p HD streams. ATSC 3.0 is a new industry standard likely to be adopted in the next 2-3 years that would use modulation to more efficiently distribute the data, at 2 bits/Hz vs. currently 0.93 bits/Hz. This would accelerate the ability to channel share and allow multiple 1080p HD streams on a single 6MHz channel.
(a) TRCO’s network agreements state it must stream at 1080p at its CBS and Fox affiliates. There are no stipulations in their CW affiliation agreements. They can likely channel share in the markets listed below but Indianapolis is not included in the total since they have no CW/My station but a Fox and CBS (this will change with ATSC 3.0 but for now I assume it does not). The chart below assumes the same values per 6MHz spectrum sales as discussed above.
(b) Channel sharing with competitors is more complicated though KLCS and KJLA announced they will channel share and participate in the auction after their successful network testing in LA. For the same reasons discussed above, I assume network sharing works only with CW/independent channels and obviously we cannot include the duopoly sharing markets tallied above in the total. I also assume they must split the proceeds with their partner, so all proceeds are halved.
(3) This has the best tax ramifications as it’s treated as a like-kind exchange which means a tax deferral, according to the FCC. The FCC has said this conversion would yield 67-80% bid prices of selling a station. I don’t really understand from where those proceeds would come, especially at the FCC’s lofty opening bid assumptions. Instead, I assume the payment will cost from the difference between net and gross auction proceeds, hence I discount my spectrum fair values by 60% (vs. 20-33% by the FCC). This yields the following proceeds which are likely tax deferred given the 1031 exchange. The FCC has stated that the satellite/cable subscriber losses are minimal in the conversion though this is a question for management (see Management Questions). Also, while VHF has worse over air range, a small percentage of TRCO subscribers still use over air antennas so even if the range is 10% worse and 5% of their viewers are over the air, this change would only affect viewership by 0.5%. One issue here is knowing how many VHF channels are available over the air, i.e. will there be enough to switch to (see Management Questions).
(4) Broadcast LTE will offer a single broadcast stream to multiple mobile devices using a single broadcast spectrum channel/tower. Broadcast LTE offers the broadcasters to lease out their spectrum and efficiently transmit live video feeds (ie heavily watched sports events), mobile OS software updates (networks are very congested when the new version if iOS is pushed to devices), breaking news, public safety alerts…etc. Broadcast LTE is neither 2-way nor very flexible but it’s extremely efficient for transmitting the same feed to many devices. The links below from Qualcomm and Sinclair offer an overview of Broadcast LTE but the highlights are that in a few years, 70% of mobile data traffic will be video, the efficiency of a point to multi-point single downlink stream is about 7x on unicast when 5 users access the same content at the same time and the limited spectral interference on broadcast networks means longer ranges. Assuming TRCO can use 3Mhz of spectrum in several years as 100% downlink on all the towers for an operator in a single DMA (NY, for example has 2,800 towers), at 3x normal spectral efficiency (lower than the 7x Qualcomm mentions), we can get to a number of GBps they can transmit. Assuming only 4% of that is ever used for unicast and $1.25/GB in leasing rates to the carriers (some MVNOs pay $2/GB today), the value of the top 7 markets at 10x FCF are listed below. Again, this is not very precise as Broadcast LTE is in test phase only but Sinclair believes it is 2-3 years away and that there’s a lot of value to this, otherwise the carriers wouldn’t want their spectrum. For conservatism, I assume this is only a viable business in the top 7 DMAs and for the reasons discussed above, it can only be shared with CW/My stations.
Real Estate Value
Tribune’s real estate assets provide near and longer term free cash flow opportunities. Near term, they can lease out currently unused square footage to retail and corporate tenants and longer term, they can redevelop sites where they are zoned for higher square footage. See appendix for detail, but I believe they can lease out an additional 1.7m square feet to generate incremental rental income near term and that the redevelopment opportunities at Tribune Tower, Times Mirror Square, 700 West Chicago, Las Olas, Costa Mesa and Orlando offer an incremental asset value. I also believe that their broadcast towers offer hidden value. Broadcast tower sales in the past have yielded $2m-3m in value (less for radio towers and more for ‘tall’ broadcast towers). I believe TRCO has about 85 broadcast towers that they can sell and lease back.
Fair Value Estimates
Valuation – Based on the targets above, TRCO should generate $4.50-5 in FCF per share pre-buybacks and then can use the cash to shrink the equity base over time, valuing an average of 2017/2018 FCF/share post buybacks at $55 at an 8.5x FCF multiple. From there and based on the analysis below, the real estate is likely worth another $9.76/share. While the spectrum scenarios are somewhat harder to assess, I probability weight the potential scenarios which the company can use to buy back stock. Assuming a best case scenario, the upside is potentially 108.7%.
Doomsday Scenario – Advertising falls off a cliff and you dump the CW stations for their spectrum value. Assuming spectrum values double between now and then, you currently have $18 in value from CW spectrum and if that triples, it’s worth $36 which provides a margin of safety but that assumes very aggressive spectrum assumptions.
Appendix Spectrum Value
FCC numbers came from October 2014 and February 2015 Greenhill slide decks. AWS-3 numbers from AWS auction, 10MHz blocks, market by market. 10MHz since that will be the block size per the FCC. My numbers below.
Assuming a 10mhz block, paired (ie 50% downlink) and normal LTE 3.7 bits/herz, 3 radios per cell site yields 55.5Mbps in capacity. At the national average of 57k cell sites, lighting up a 10mhz block would yield 3.182mln Mbps. Assuming your average carrier has 50mhz of LTE spectrum right now, creating the same amount of capacity (3.182mln Mbps), they would need an incremental 8,602 towers, ie to expand their network capacity by the equivalent of 10mhz of spectrum would require building on 8,602 new towers. I then list out the costs associated with building on new towers and subtract the ongoing costs associated with lighting up the new spectrum. I then divide this by the population to get to $2.24 per Mhz-Pop. For each individual market, I substitute its own population, number of towers and tower lease rates. We have the DMA populations. I calculate the number of towers as one per 400 sq miles (coverage) and an additional tower per each 7k people (capacity). This checks out with estimates of towers for the total TRCO DMA areas. This yields my estimates for MHz POP valuations used in the tables above.
(1) Participate in the auction and close down stations
From lender decks, we know 2012 revenue and OCF per legacy TRCO station
We have estimates for recent and past retrans revenue per sub
From here, I back out retrans revenue per station (we have some sense of reverse comp costs from old lender deck) to look at non-retrans OCF and estimate revenue/margins per LTV station (revenue per sub is much lower on CW/My stations but margins are higher)