Monday, January 16, 2023

Proximus Capital Markets Day, strategy 2023-2025 'bold2025'

Proximus CMD: Strategic plan 2023-2025 bold2025 (press release, presentation)


  1. Best gigabit network:
    • FTTH coverage 50% by 2025 (>3m HP with >1m HA = 45% of customers vs 55% on copper, 500k exclusive customers by 2025), 95% (6m HP) by 2032
    • Target copper-free by 2035, savings on power & maintenance max 120m EUR/annum, avoided renewal max 130m EUR/annum (together 250 EUR/annum), max 230m EUR/annum for upgrade)
    • Current copper-to-fiber migration 50% after 6 mo, market share gain 2 pp after 12 months; churn 30% lower, repair cost per customer 40% lower; fiber pricing 5-12 EUR/mo over copper, 30 EUR/mo over copper for 10 Gb/s tier, ARPC uplift 7 EUR/mo before promos (4 after)
    • Currently in >90 cities adding 600k lines/annum, current Unit Cost EUR 940, to grow to 950 post inflation & savings
    • New tech on fiber: quantum channel (for encryption), slicing (for gaming), digital twin (for monitoring), 25G-PON
    • 5G nationwide by 2025 (to apply slicing; 3G phase-out end 2024)
  2. Upgrade IT to support digital ecosystems (convergent solutions, shift to e-sales & e-servicing) and save costs (TCO reduction EUR 70m by 2025 in opex & capex)
  3. #thinkpossible culture, agile methods
  4. Inclusive society, protect environment (truly circular by 2030, net zero value chain by 2040), close the digital divide
  5. Best customer experience by 2025: digital-first, Picks (aggregator = asset-light with select exclusives), new Proximus+ service (TBA, target 1.8m active users by 2025) bundling daily services: telecoms, fintech (neobank), mobility (in 1 app), e-health (telemedicine), energy (innovation)
  6. Grow Proximus Domestic (fiber, multi-brand (Proximus (premium, family), Scarlet (value), Mobile Vikings (innovators, cord cutters)), FMC, ICT) & International (BICS, TeleSign)


  • Domestic:
    • rev growth 2022 2%, 2023 1-3%
    • EBITDA 2022 +1%, 2023 -3% on inflation, EBITDA to grow from 2024, EBITDA 2025 to equal 2022
  • Group:
    • EBITDA 2022 +1%, 2023 -3%, EBITDA 2025 slightly higher than 2022
    • new cost (opex) savings plan EUR 220m over 3 yr (o/w 40% from workforce, 30% from IT, 30% from network/energy)
    • capex peaks in 2022 and 2023 at EUR 1.3b (fiber capex 20% on balance sheet, 80% off)
    • to divest non-strategic assets to raise EUR 400m (incl 143m from CHQ, rest from infra & property, excludes option to sell mobile towers and BICS, TeleSign)
    • dividend 2023 EUR 1.20, reduced to rebased level from 2024 and 2025 to EUR 0.60
    • to return to FCF growth from 2024
    • leverage 1.6x (excl. off-balance), as defined by S&P 2.3x (to 2.6 in 2023, 2.5-3.0x during 2022-2025)
  • International:
    • rev growth HSD, combined rev EUR 1.8b by 2025
    • direct margin CAGR 2022-2025 HSD

Thursday, January 12, 2023

Fair Share Contribution (aka Internet Traffic Tax) revisited - Big Telco vs. Big Tech

There are some problems with Big Telco's (ISPs) reasoning claiming that Big Tech (CAPs) should contribute their 'fair share' of broadband investments.

1. Net Neutrality

Applying the Internet Traffic Tax (ITT) to Big Tech only would violate Net Neutrality, meaning: it would create barriers to entry and growth for smaller and new CAPs, as Analysys Mason argues (and goes on to state that CAPs could have reduced incentives to invest in infrastructure, which would increase costs for ISPs, risk reduced quality of internet access, reduced ISP competition and ultimately higher fees for end-users).

By the way: what if the most popular services were not clustered at a handfull Big Tech companies, but instead were widely spread over hundreds or even more companies? What would the ITT then look like?

2. Fair is fair

Co-investment would necessarily lead to Big Tech taking an equity stake in a series of infrastructure joint-ventures with Big Telco.

Unless it would indeed be a tax, from which a European Broadband Fund would be funded, but that would alter the Big Telco proposal entirely.

Conversely, if Big Tech should share in the cost of access networks, then Big Telco should share in the cost of developing services and content.

Indeed, the EU appears to be signalling that the cost of for instance the metaverse and the cloud (incl. subsea cable systems and datacenters) should be considered - areas where Big Tech's investments dwarf those of Big Telco.

3. If it ain't broke, don't fix it

Increased data traffic leads to increased network costs and investments. This is a natural consequence of Big Tech services leading to the creation of the (large, growing and very profitable) Broadband Market in the first place. However, over the past few years Big Telco margins have only expanded, as the cost per bit has been coming down.

Peering and local caching (such as Netflix's free Open Connect CDN) only help towards this. The system, based on good old market forces, works fine. Some, such as DT and SKT, resist Open Connect, leading to more expensive transit (in the case of SKT: to Hong Kong and Tokyo), deteriorating the user experience.

4. Two-sided business model

Big Telco can't be forced to adopt a network vision including peering, local caching, transport (incl. subsea) and hosting (datacenters) and appears to be more interested in creating a two-sided business model based on Sending Party Network Pays (SPNP), as ETNO proposes (and BEREC opposed). Even if the former (peering and caching) provides Big Telco with large savings, as Analysys Mason shows. SPNP carries the risk of giving Big Telco monopoly power in termination, as Euro-IX argues.

Big Telco owns the billing relation with the end-user. If it has insufficient pricing power to raise prices, then this implies that not all Telcos agree with Big Telco's argument. Also, why not switch from unmetered to metered (volume-based) tiers? And remeber: manufacturers of electronic products never contributed to the cost of the electricity grid.

Tuesday, December 20, 2022

Outlook 2023: 9 predictions for the Dutch market

1. The multi-gig is not embraced en masse

For now, Delta is the champion of the multi-gig offering (8 Gbps). However, the uptake of services above 1 Gbps is quite low, although it can vary greatly regionally. It seems to be more for the business market for now, although something like the metaverse could also drive demand for bandwidth for consumers. We do not expect KPN and Ziggo to feel the pressure to participate in this early phase of a new trend. It is technically difficult: KPN's XGS-PON network is far from national and Docsis 3.1 requires adjustments to go above 1 Gbps. Even the challenger T-Mobile apparently has no plans. It will be after 2023 before we start purchasing the multi-gig on a large scale.

2. Delta Fiber buys Open Dutch Fiber

If investors have to answer the question 'what is a long-term investment', the answer is 'a failed short-term investment'. This also threatens Open Dutch Fiber. It seemed so easy: quickly install fiber optic networks in the major cities, withdraw customers en masse from KPN (and Ziggo) and then put KPN in front of the block to take over the company. It turns out not to be that easy. KPN builds just as quickly and for less than 1,000 euros per line. KPN will reject a takeover for a multiple of this multiple. Furthermore, it relies on its brand name and offerings to win back customers.

Delta Fiber is different in the competition. There is no overlap with Open Dutch Fiber and a merger is obvious with a deal in shares. EQT, Stonepeak, KKR and DTCP share ownership and with certain rights KKR, for example, can be awarded an elegant exit.

3. T-Mobile faces competition on Open Dutch Fiber

Well, this is an open door, but ODF will eagerly await the arrival of multiple ISPs on its network. It is wise if you have strong infrastructure competition, such as Delta Fiber in its home country Zeeland. According to insiders, the deal with T-Mobile is that the latter has exclusive rights as ISP on every network for a year. That period will end first in Zoetermeer. We count on providers such as Online, Youfone, Kliksafe, Freedom Internet and perhaps even Solcon (KPN) to be eager to join from the beginning of 2023.

4. No fourth MNO

New technology can turn an industry upside down. Fiber optic has generated a large number of new entrants, sometimes locally or regionally, and sometimes with a new business model, such as wholesale-only. Unfortunately, the mobile sector is much more locked down. After the last consolidation (T-Mobile bought Tele2 and Simpel) the number of MNOs went back to three and the largest MVNO disappeared.

In other countries, after such a round, there appears to be room for a newcomer and fourth MNO, such as in Italy (Iliad) and Japan (Rakuten), and soon in Germany (1&1 from United Internet), Belgium (Citymesh Mobile from Cegeka and Digi) and the US (Dish). They rely on various things for this: affordable spectrum, access to a wholesale provider of installation points (such as Cellnex in the Netherlands) and relatively cheap cloud technology (such as that of Rakuten, which is also used by e.g. 1&1).

No one in the Netherlands sees this gap in the market.

5. VodafoneZiggo is selling its pylons

Last May, parent company Liberty Global announced that it was considering selling VodafoneZiggo's mobile sites. It has not come to that, but with major investments on the way, it is now more obvious. The debt position cannot go up any further and if the dividend cannot be reduced, then there is only one option: to sell the 'silver'. Half-sister Telenet preceded VodafoneZiggo in this, but interest rates have since risen further. This means that the buyer cannot so easily load the mast company with debt.

Sales could generate at least a billion euros, money that is badly needed for the 5G auction of Q3 2023 and the content ambitions (interest for the Eredivisie).

6. T-Mobile discontinues its television platform

The television market is changing rapidly. With the arrival of SkyShowtime, a provisional endpoint has been reached where a large number of apps (Netflix, Prime Video, Disney+, HBO Max, SkyShowtime, Apple TV+) compete for the favor of the viewer. New hardware (smart TV, casting devices) and software (such as Google TV) reduce linear television to 'just another app', supported by NLziet's stand-alone OTT-TV apps (NPO, RTL, Talpa for EUR 8/month) and Canal Digital (an even wider range for EUR 15/month).

For now, broadband customers usually ask for the double play BB + TV but that is shifting. Cord cutting will continue to increase. This makes it expensive to maintain a television platform. The ISP can simply refer to the mentioned hardware, software and streamers. T-Mobile in particular can show itself to be a forerunner in this. An alternative is the Zattoo platform service.

Maybe 2023 is too soon, but T-Mobile wouldn't be the first to take this step.

7. Stacking increases

The SVOD providers mentioned above almost all have a qualitatively and quantitatively excellent offer. With OTT-TV as 'just another app', cord cutting is increasing and budget is freed up for a new subscription. Unless the economic malaise throws a spanner in the works, there will be room for a new subscription.

8. Viaplay buys the Eredivisie rights

The rights to the Eredivisie will soon be released and will be available for a new period of seven years. The return of Bob Iger as CEO of Walt Disney (parent company of ESPN) suggests that costs are being scrutinized. ESPN Netherlands is probably not a core activity, which increases the chances of the rights being transferred to another party. According to reports, Ziggo is a candidate, but the costs may be too high, reportedly EUR 150 million per season. Then Viaplay is the obvious choice.

9. RTL-Talpa merger called off

Apparently, the authorities (ACM) and the candidates are busy assessing the merger plans of RTL Nederland and Talpa Network. Undoubtedly, concessions play a major role in this, because if France is leading, then the omens are unfavorable.

The big question is whether RTL and Talpa can continue the success of Videoland by working together without a merger. There is still considerable growth, but with the arrival of the major American parties, Videoland's market share is decreasing slightly. If they think they need a full merger for this, then it suggests that they also need the other benefits of the merger in order to survive (they're already losing UEFA rights too): cost benefits from eliminating overlap and revenue benefits from a stronger position vis-à-vis producers and advertisers. However, the latter is a 'no go' for ACM. ACM wants to protect the positions of producers and advertisers.

If RTL and Talpa opt for a limited collaboration in Videoland, the next question is what will happen to NLziet. In France, commercial broadcasters are withdrawing from Salto, comparable to NLziet, now that their merger has been banned. That would mean the end of NLziet. Perhaps that gives the parties extra leverage to push through their merger with ACM: "if we are not allowed to merge, we will stop NLziet".

Originally published here (in Dutch)

Friday, December 09, 2022

Walt Disney launches Disney+ Basic in US

Launches ad-supported tier Disney+ Basic in US

  • 8 $/mo
  • ad-load 4 min/hr (no ads on kids profiles & preschool content; >100 advertisers at launch: Google, P&G, Verizon, Taco Bell, Walmart, etc)
  • all content incl HD
  • max 4 simultaneous screens, max 7 profiles
  • no downloads
  • raises ad-free tier from 8 to 11 $/mo or 110 $/yr
  • bundle prices:
    • Disney+ & Hulu with ads (Duo Basic) 10 $/mo
    • Disney+ & Hulu & ESPN+ with ads (Trio Baisc) 13 $/mo
    • Disney+ & Hulu With Live TV & ESPN+ with ads 70 $/mo
    • Disney+ & Hulu & ESPN+ ad-free (Trio Premium) 20 $/mo
    • Disney+ ad-free & Hulu with ads & ESPN+ 15 $/mo only for existing subs

Tuesday, December 06, 2022

AWS re:invent 2022

AWS re:Invent 2022 (221128 - 221202, Las Vegas):

  • No hiring freeze, keeps building datacenters
  • Plans to be water positive (water+) from 2030 (returning more water to communities than it uses)
  • Signs Descartes Labs; and Wallbox; and Brookfield AM; and Stability AI; and Yahoo; and American Family Insurance
  • Vodafone, Intel, Dish Wireless, Swisscom, Spark, Telenor, Telef√≥nica, T-Systems, JMA Wireless, others demo MEC & RAN solutions
  • Expands partnership with Slalom to develop vertical solutions and accelerators on AWS for customers in the energy, financial services, healthcare, life sciences, public sector, and media & entertainment industries - Expands partnership with Atos (enables Atos customers with large-scale infrastructure outsourcing contracts to accelerate workload migrations to the cloud and achieve digital transformation)
  • Launches tools to connect & analyse data stores - Launches Amazon DataZone - Launches Amazon Security Lake - Launches AWS SimSpace Weaver - Launches 3 new Amazon Elastic Compute Cloud (Amazon EC2) instances powered by 3 new AWS-designed chips - Launches Graviton3E chip - Launches AWS Supply Chain - Launches AWS Clean Rooms - Launches Application Composer (low-code tool for builduing serverless apps) - Launches  AWS AI Service Cards
  • Adds 5 new capabilities for Amazon QuickSight - Adds 5 new Database & Analytics capabilities - Adds 8 new Amazon SageMaker capabilities

Thursday, November 10, 2022

United Internet's and 1&1's 5G roll-out plan updated: FWA first

1&1 (United Internet) presented a new time table (slide 10) for its 5G network roll-out:

  • FWA launch 22Q4
  • Nationwide marketing launch, and end to MVNO sales, 23Q3

Wednesday, October 19, 2022

Netflix Q3


  • "After a challenging first half, we believe we’re on a path to reaccelerate growth. The key is pleasing members."
  • "We think our bingeable release model helps drive substantial engagement, especially for newer titles."
  • "Our existing plans remain ad free."
  • To stop providing subs forecasts from 23Q1
  • Strong USD impact 2022E $1b on rev, $0.8b on oper income; oper margin target (19-20%) on track but lower if USD remains above Jan 2022 level
  • Targets FCF: 2022 $1b (+/-  few 100m), 2023 substantial growth
  • Growth rates excluding currency effects:
    • 22Q3
      • Total streamers (223m) +4.5% (lowest ever)
      • ARM ($11.85) +8%
      • Revenues ($7.93b) +13%
    • 22Q4E
      • Total streamers (228m) +2.6%
      • ARM +6%
      • Revenues ($7.78b) +9%
  • Margins
    • Gross margin (after content amortisation) 39.6% (down from a peak 21H1).
    • Marketing expense 7.2% of revenues (roughly at a low)
    • Technology & Development roughly flat at ~8.5% of revenues
    • Operating margin 19.3%
  • Cash flow
    • Cash spent on streaming content $4.52b in Q3, $17.8b TTM (stable for 4 quarters)
    • Cash spending-to-amortisation ratio 1.24x (peaked at 1,75x in 19Q4)
    • Cash spent per net add $1,884 (peak)
    • Cash spent per retained sub $20.4 (longer term roughly flat)
    • Cash & equivalents $6.11b. Last debt issue 20Q2, then bottomed at $5.82b in 22Q2.
  • "We now have 35 games on service (all included in every Netflix subscription without in-game ads or in-app purchases) and we’re seeing some encouraging signs of gameplay leading to higher retention." + 55 games in development
  • Animal Logic acquisition [see 220719] to impact 22Q4 cashflow
  • Plans new game studio in S California
  • Considers cloud gaming service
Password sharing
  • To start charging sharers extra for borrowers from early 2023.
  • Borrowers may create own subscription, perhaps the Basic With Ads tier.
  • Borrowers can migrate their profiles using Profile Transfer.
Main points
  • Back to subscriber growth.
  • Heavy USD impact.
  • Focus on customer: content (originals) & binge viewing.
  • Games becoming more important.
  • Password sharing attacked with charging plan, cheaper tier and Profile Transfer.
  • Margins could be expanded, as scale builds, but games delay the process.
  • Content spending stabilising in absolute terms.

Monday, October 17, 2022

Vodafone selects Altice for open access fiber joint venture

Vodafone selects Altice for open access fiber joint venture (instead of KKR, Brookfield or Deutsche Glasfaser/EQT).

  • Vodafone DE to establish 50/50 JV FibreCo 23H1 with Altice
  • To develop 7m FTTH lines (o/w 80% to housing ass in existing HFC footprint, 20% to neighbouring homes outside current footprint) in 6 yr
  • Total investment EUR 7b o/w 70% debt financed
  • Currently offering 1 Gb/s to >24m homes (incl wholesale on Telekom, ...); to complement node splitiing, Docsis 3.1 (high splitting, max 3 Gb/s), Docsis 4.0
  • Open for wholesale (Vodafone DE anchor tenant without minimum revenue or volume commitment)
  • Construction & maintenance by Geodesia (= Altice)
  • Vodafone DE to receive max EUR 1.2b cash (o/w 120m upfront at closing JV, 487m deferred during roll-out, earn-out max 595m)
Current JV's in Germany:

Friday, October 14, 2022

Netflix's tier #4 coming November 1, 2022: Basic With Ads - Updated

To launch AVOD tier Basic With Ads in 12 countries from November 1, 2022:

  • 221101 in Canada (6 CAD/mo), Mexico
  • 221103 in US (7 $/mo), UK (5 GBP/mo), Australia (7 AUD/mo), Brazil, Japan, S Korea, DE (5 EUR/mo), FR (6 EUR/mo), IT (5.5 EUR/mo)
  • 221110 in ES (5.5 EUR/mo)
  • To provide outlook at 22Q3 (221018), questions:
    • Growth (take-up) prospects
    • Customer satisfaction
    • ARPU impact
    • Cannibalisation. How many will sign-up as new customers vs. current customers downgrading.
    • CPM (current and future, assuming stronger targeting)
    • Will there also be, in time, a 'Standard With Ads' and a 'Premium With Ads' tier?
  • "We're confident that ... we now have a price and plan for every fan."


  • 1 simultaneous device
  • HD/720p
  • No downloads
  • Excludes 5-10% of titles as result of rights restrictions (to be reduced over time)


  • Pre & post roll ads, new movies pre-rolls only
  • 15-30 sec (20 sec in ES)
  • Ad load 4-5 min/hr
  • Targeting at first based on Top 10 by region & genre, later possibly on age, gender, viewing behavior, time of day.
  • Ads sales based on fixed price not auction.
  • Nielsen’s Digital Ad Ratings (DAR) service to start measuring 2023, eventually through Nielsen ONE Ads.
  • DoubleVerify and Integral Ad Science (IAS) to verify viewability & traffic validity of ads from 23Q1.

Comparison to current tiers:

  • To upgrade Basic plan to HD (720p).
  • Current US tiers Basic, Standard, Premium 10, 15.5, 20 $/mo
  • Current US DVD-by-Mail: 10, 15, 20 $/mo for 1, 2, 3 discs concurrently

Monday, October 10, 2022

Why FWA is serious business in Denmark, but not in The Netherlands - Updated

  • DK and NL have high levels of fiber.
  • The 3.5 GHz band was auctioned in DK in 2021, NL will follow 2023.
  • 3 Denmark (Hi3G DK) embarked on heavy investment in network desnification (see graph). They are now marketing FWA-over-5G as an alternative to fiber.
  • MNOs in NL will refrain from this kind of investment: 1. VodafoneZiggo and T-Mobile NL are heavily leveraged, they need to pay large annual dividends and they will stay away from any serious market disruption, 2. In the mobile oligopoly, KPN also pays out large dividends and will not feel challenged.