Description

O6I has a fantastic, in-depth write-up of the business, which I highly recommend reading (if you haven’t already). Link here: https://valueinvestorsclub.com/idea/TENCENT_HOLDINGS_LTD/4356829474

Rather than rehash much of what was covered with respect to the business, the focus of this write-up is “why now,” and where we believe there to be a misunderstood opportunity.

SUMMARY

After two years of regulatory-induced pressure, Tencent’s gaming business is set-up for a likely banner year in 2020, with indicators that this is already underway, and several titles appear to be underappreciated by the market, both with respect to timing and addressable market opportunity. 

Advertising revenue appears to be rebounding after a soft 2019, in part due to macro concerns at the beginning of the year, but also due to a change in rev recognition. While it is widely expected that advertising will be strong in 2019, the magnitude of the recovery and margin trajectory appear incorrectly reflected in consensus numbers.

It appears Tencent has reached an inflection point in its FinTech business, with several near-term revenue and margin tailwinds, driven by increasingly rational competition, scale, and having cemented itself into consumer behavior driving long-term sustainability. 

While there are three core drivers that appear underappreciated, which will likely drive outperformance over the next 1-2 years, Tencent is a high-quality business with sustainable competitive advantages, which should compound earnings power over the long-term.

BUSINESS OVERVIEW

Tencent operates what began as a messaging app, but is now the Chinese ubiquitous platform / mobile operating system that is deeply integrated within the Chinese internet consumer’s daily life. Tencent effectively “owns” the Chinese internet user (~1.2bn WeChat MAUs as of 3Q19), which it then monetizes primarily through video games, advertising, and payments. It runs one of the largest payment networks in China, and it is the largest gaming company in the world. The business has one of the widest moats amongst global internet companies, with an ecosystem of integrated products and features driving internal and external network effects. It also benefits from regulatory barriers in nearly all its business lines. The durability of its moat is demonstrated by consistent >100% ROIC (assuming 10% of sales of operating cash and excluding book value of investments) prior to the 2018 gaming freeze. Regulatory risk is tempered by the fact that Tencent, in many ways, administers the most comprehensive and powerful surveillance network in all of China (the unencrypted, censored & monitored WeChat).

THESIS DETAIL

1. After two years of regulatory-induced pressure, Tencent’s gaming business is set-up for a likely banner year in 2020, with indicators that this is already underway, and several titles appear to be underappreciated by the market, both with respect to timing and addressable market opportunity.

One of the primary methods Tencent employs to monetize its user base is video games. And as one of the largest game distributors, it has more access to consumer behavior data, which informs development. This dynamic grants Tencent a competitive advantage, which is further strengthened by regulatory barriers. International game developers need to work with local partners in order to launch games in China. Tencent is an attractive partner given its massive user base (essentially all Chinese internet users), which forms a steady inflow of gaming titles. 

Without delving too deep into minutiae (as we assume much is already understood from O6I's write-up), in April 2018 new game licenses were suspended during the Chinese government’s reshuffle. Changes were made to the agencies that have control over the approval process. At a high level, more control has shifted to the State propaganda department. The reshuffle was completed in March 2019, and approvals were resumed. However, a second game license approval freeze was implemented on any new games in order to give the regulators time to work through the backlog of the title submissions from the prior freeze. It took approximately 9 months to work through the backlog, and we are now largely in a steady-state regulatory environment. 

Given the two years of regulatory pressure, Tencent’s margins and return on capital suffered, as they continued to invest in development for the future without the ability to capture any of the associated returns. Essentially, gaming returns on ~2H17-2H19 invested capital were delayed by approximately 2 years.

There are several titles that will likely launch throughout this year that appear to be underappreciated by the market. The market has focused on DnF Mobile and its potential given the success of its PC counterpart (approx. ~20-25% of 2019 PC gaming sales, and consistently ranked worldwide as one of the top 2 grossing PC titles in 2019, per Superdata). The game is expected to be launched in 1H20, and there is uncertainty around its current status with respect to regulatory approval. Per Douyu (China game livestreaming company), it has already been conditionally approved after minor refinements. Per Bytedance gaming mgmt, it has already been approved. The game has already been through several rounds of beta testing, and pre-registrations opened at the end of 2019 with over 10mn registrations as of Jan 3rd. There is a modest probability DnF Mobile launches in 1Q20, however we assume a June launch in our base case (see charts below for base case assumptions). 

What the market appears to be missing, or in some cases only now beginning to appreciate, is the revenue opportunity of several games to be launched this year. 

Conversations with eight industry participants (gaming companies, professional eSports teams, and developers), unanimously believe that one game, LoL mobile, will become a global mega-hit with staying power. League of Legends PC, like DnF, was consistently ranked worldwide as one of the top 2 grossing PC titles in 2019, per Superdata. 

China game livestreaming companies, Huya & Douyu, both have shared that they have beta accounts, and after internal tests and initial feedback, they believe LoL mobile, along with DnF Mobile, will be the biggest incremental drivers to their business this year.  Conversations with Tencent’s gaming department indicate that they hold the same belief. 

Additionally, a limited number of PC-based LoL Esports players were invited to test Wild Rift. Several of these testers from renowned LoL esports teams noted a very high level of enthusiasm for the game. After testing, teams are already preparing for competitive play, which in of itself is a meaningful user acquisition channel. Tester feedback has been overwhelmingly positive. For example, one tester described a series of complicated maneuvers professional esports players use in the desktop game that, to his surprise, was perfectly replicated in the mobile version. 

To summarize, contrary to the many times games with high revenue expectations launch but are subsequently met with a lukewarm user response, multiple data points indicate the exact opposite is likely to happen with LoL Mobile (which shouldn’t come as too much a surprise, given the developer of the game, Riot, is obsessed with user experience). 

Douyu shared that their understanding is that Tencent has already received preliminary regulatory approval for LoL mobile. Bytedance gaming operations said the same. 

The official target launch is 2020.

In addition to LoL Mobile, Riot games is scheduled to also launch several other significant titles in 2020. 

• Teamfight Tactics (TFT) Mobile, which like DnF, is an already successful PC game. Developer Stephen Mortimer & Dax Andrus announced on Jan 13 that TFT Mobile will be coming mid-March.

• Legends of Runeterra, which is similar to Blizzard’s Hearthstone, is slated for a Jan 23rd launch. Of note, the worldwide trading card game market was approx. $1.8bn in 2019.

Lastly, Call of Duty Mobile is likely to launch this year, per industry conversations. Its approximate annual revenue run-rate (ex-China) is ~$200-300mn. Bytedance noted that Tencent has already received soft approval for this game as well.

The conversations referenced above informed our base case bottoms-up gaming build, yielding 2020/2021 gaming rev estimates +5%/+8% above consensus. See charts below.

2. Advertising revenue appears to be rebounding after a soft 2019, in part due to macro concerns at the beginning of the year, but also due to a change in rev recognition. While it is widely expected that advertising will be strong in 2019, the magnitude of the recovery and margin trajectory appear incorrectly reflected in consensus numbers.

Exiting 2018 and in 1H19, Tencent’s advertising business was hit due to advertiser macro concerns. At the same time, Tencent quietly changed its accounting standard, recognizing a portion of ad revenue on a net basis vs. gross prior. The result was a sharp deceleration in growth from +44% y/y in FY18 to +13% y/y in 3Q19. 

However, during this same period, due to the change in rev rec, incremental Gross Margins in the advertising business expanded from 34% in 2018 to 141% in 3Q19. 

Conversations with one of Tencent’s largest advertising partners (iClick) and Bytedance indicate that advertisers pulled back spend due to macro uncertainty in 1H19, began cautiously redeploying budget in 3Q19, but gained confidence in August/September and found themselves sitting on a substantial unused amount of their 2019 allocated annual budget. Advertisers and ad tech providers in Beijing, Hong Kong & Shenzhen noted that advertisers began a rush to spend the unused budget in 4Q19 so that come time to budget for 2020, they wouldn’t be pegged to a lower number. Each conversation we have had suggest a sharp reacceleration in ad spend in 4Q19, and while FY20 budgets aren’t completed until following Chinese New Year (and now pushed out due to the coronavirus), preliminary partner conversations indicate expectations for Tencent social ad spend to accelerate in 2020, suggesting >32% y/y growth vs. consensus expectations of +27% y/y growth. Particularly of note, ad load in WeChat moments has gone from 2 per day to 3 per day, with a 4th being tested.

Additionally, consensus expects incremental Advertising Gross Margins of 48%/46%/46% in 2020/2021/2022 vs. ~114% in 2019E. Factoring in the mix-effect from lower-margin media advertising (e.g. video ads), we estimate 2020/2021 Advertising GMs of 50%/51% vs. 49% in 3Q19, ~300/400bps above consensus.

3. It appears Tencent has reached an inflection point in its FinTech business, with several near-term revenue and margin tailwinds, driven by increasingly rational competition, scale, and having cemented itself into consumer behavior driving long-term sustainability.

The payment process in China is relatively simple compared to the US, with effective take-rates (effective Merchant Discount Rate, or MDR) depressed over the last decade as Tencent & BABA have fought to gain market share & drive market-wide penetration. 

To briefly summarize the process, in China the payment process is typically facilitated by 5 parties. A merchant acquirer (e.g. Huifu, Yeahka, Lakala) provides merchants with a POS machine and connects the merchant with the payment network. The consumer has an issuer account, their bank or third-party payment account (e.g. WeChat Pay / Alipay). And the clearing network (e.g. Union Pay) facilitates the clearing process. 

Over much of the last decade, the game was TenPay/Alipay offering free QR codes to merchants & acquirers and then subsidizing its use through discounts to both the merchant (paid typically as a lump sum after a certain value threshold) and the consumer (coupons). After 3rd party payment providers (the duopoly) reached >60% penetration in early 2019 (now ~80% penetration), the 2 players began to act more rationally w/ pricing/subsidies, and since mid-2019 both players have been steadily reducing subsidies. The nominal MDR is ~60bps for both third party payment networks (e.g. QR codes) and bank cards. However, conversations with two of the three largest merchant acquirers in China noted that due to discounts, the effective MDR to merchants was ~30bps in 2017, but that due to market rationalization, the effective MDR has increased to ~40bps at the end of 2019 and is expected to continue narrowing the effective and nominal gap (the remaining 20bps). 

The rational pricing behavior by the third party payment providers is being driven in part by penetration, but also largely due to consumer behavior / habit. With respect to Tencent, the merchant acquirers explained that starting after sending red packets via WeChat Pay became the norm, users would default to WeChat pay despite higher rebates/discounts/coupons from AliPay. But different payment methods have different use cases. E.g. it is increasingly becoming the norm for Apple Pay or Huawei Pay RFID transit pay be used for the train (1 tap entry), but when consumers visit a merchant they “forget about [ApplePay / Huawei Pay]” and immediately default to WeChat Pay “without thinking.” One merchant acquirer summed it up best: “the habit formation is crazy.” There is a significant behavioral / habitual element here. It brings to mind Buffet’s comments on Coke – the competitive advantage doesn’t lie with the brand or price, but consumer behavior & habit. There are also significant textbook network effects even beyond those of Tencent’s core social media business (though it’s more a portal / mobile operating system). While building a social network is difficult, where the obstacle is getting people to enter an email address and password, WeChat Pay is similar to Visa/Mastercard in that it distributes a means of accepting payments to merchants and a means of making payments to consumers.  And it is stronger in that it is built on its entrenched ecosystem of users (i.e. largely the China internet population). WeChat Pay, more so than Alipay, arguably has one of the widest moats among global internet businesses today.

The trend toward increasing net take-rates has already begun and will continue, per conversations with merchant acquirers, driven by Tencent’s push to acquire merchants directly (cutting out the ~15-20bps paid to the acquirer), and as reflected by Tencent’s FinTech margins inflecting positively in 2019 from breakeven in 2H18 to double-digit operating margins in 3Q19. Notably, all three business lines within FinTech (payments, wealth management, and consumer loans) have positive operating profits. Additionally, beginning this year, the PBOC will begin allowing third-party payment providers to accrue interest on their customer deposits at 35bps pa, effective beginning Aug 1, 2019, and lasting 3 years. We estimate the incremental escrow interest will contribute ~150-200bps to segment margins. 

These trends do not appear to be reflected in forward consensus numbers. We estimate 2020/2021 Fintech & Business Services GMs ~380/760bps above consensus. This implies incremental GMs of 43%/49% in 2020/2021 vs. consensus 29%/30%, and compared to 35% incremental GM in 3Q19.

In short, WeChat Pay is a remarkable business that is inflecting positively, with upside to consensus estimates over the next 1-2 years, and the ability to compound at high returns on capital over the long run given its sustainable competitive advantages.

RISKS

• Riot (developer of LoL Mobile) is well known for having perfectionist tendencies. There is a risk that new game releases are delayed beyond our base case assumptions.

• While Alipay & Tenpay are currently acting rationally re: pricing/discounting, there is a risk that this may change if/when new payment technologies come to market (e.g. facial recognition).

• Bytedance is entering into the gaming space. However, as of now their development focus is on casual games, primarily monetized through advertising (similar to French gaming company Voodoo).

• Coronavirus impact on near-term sentiment, potential larger impact to macro (though tempered somewhat by a surge in gaming use causing several of Tencent’s gaming servers to recently crash from overload)

• Regulatory risk – China may want more control over third-party payment networks, future game releases may be denied licenses due to more stringent criteria, and the typical regulatory risks one needs to gain comfort with when investing in China internet. However, as mentioned above, we believe this is tempered by the fact that Tencent, in many ways, administers the most comprehensive and powerful surveillance network in all of China (the unencrypted, censored & monitored WeChat).

VALUATION

• Base Case: HKD 547

• Core: 20x 2021E normalized EPS excl. FinTech (includes SBC as an expense, but excludes amort of acquired intangibles & 1x gains/disposals/impairments)

• FinTech: 25x 2022E NOPAT, discounted at 10%. Assumptions:

• Third-party payment market share (excl. P2P/Social transfers) expands from ~40% in 2018 and ~44% in 2019 (per merchant acquirer & other industry convos) to ~46% in 2022

• Nominal Merchant Discount Rate (MDR) unchanged at ~60bps, but discounts decline from ~20bps in 2018 to ~5bps in 2022

• Operating margins of 26% in 2022E – informed by bottoms-up build, but reasonable when compared to comparable FinTech operating margins

• Investments: Valued at current market price or the latest transaction price with a 20% HoldCo discount

• Consolidated: Variant view on Payments, Gaming & Advertising drive base case 2020/2021 consolidated revenue 4%/6% and EPS 16%/27% above current consensus. SOTP value implies 26x 2021E EPS, 1 turn above current forward PE on consensus numbers

• Bull Case: HKD 655

• Core: 25x 2021E normalized EPS excl. FinTech (includes SBC as an expense, but excludes amort of acquired intangibles & 1x gains/disposals/impairments)

• FinTech: 25x 2023E NOPAT, discounted 2-years at 10%. Assumptions:

• Third-party payment market share of ~46%

• MDR discounts decline to 0bps

• Operating margins of 31%

• Investments: Valued at current market price or the latest transaction price with a 20% HoldCo discount

• Consolidated: Variant view on Payments, Gaming & Advertising drive base case 2020/2021 consolidated revenue 4%/6% and Non-GAAP EPS 16%/27% above current consensus. SOTP value implies 31x 2021E EPS

Catalyst

• Core gaming IP launches (Discord, Reddit & Taptap are good places to follow beta progress)

• Continued fintech margin expansion & Ant Financial IPO unlock SOTP value

• Advertising revenues reaccelerate following the resumption of content approvals & lapping change in rev rec