Description

Disclaimer: I and/or others I advise hold a material investment in the issuer's securities. Nothing here constitutes a direct solicitation/recommendation to buy/sell a security. You are advised to do your own due diligence on the security(ies) mentioned prior to making any potential investment decision.

At current levels ($57.13), I think buying the ADRs of Weibo (WB US) offers an attractive risk-reward.  I expect the stock to double over 3 years.  For a useful background (and a better pitch) on the business, please also see ahab931’s write-up on SINA from Jan 2017.  The current situation is akin to trying to catch a GARP-profile business at a time where it (and its sector) are exhibiting a falling-knife price-chart.

Intro
Almost all China large-cap and SMID-cap internet stocks are heavily selling-off in the past few quarters. In general, these businesses have the following attractive factors –
• large user bases with network effects;
• high absolute and incremental margins;
• tailwind from a still growing smartphone penetration and rising consumer affluence;
• continued ad substitution towards digital ads;
• rising time share on mobile platforms from fragmented time due to increasingly compelling content, user interface design or the use of algorithms by the platforms that make the platforms more addictive.

These coincide with cash-conversion that is enabling the companies to shovel substantial sums into R&D, capex, acquisitions, JVs and equity investments in a land-grab to lock-up footholds in various adjacent sectors, often within China but also sometimes internationally (occasionally, these moves are head-scratchers e.g. Tencent’s stake in TSLA is one example).  When sentiment and newsflow around China (and China tech) is in a positive cycle, sell-side & investors tolerate (or possibly even reward) these investments, the companies trade on their TAM and EV/sales, the delta between GAAP and non-GAAP EPS is tolerated or indirectly praised as evidence of ongoing far-sighted investments, and any technological or competitive concerns are brushed away with rationalizations that these are winner-take-all (or winner-take-most) markets where the entrenched ecosystem leaders will eventually muscle the competition out or neutralize it via acquisitions.   

But, when sentiment worsens, the focus shifts from sales & TAM to near-term GAAP earnings, sell-side & investors suddenly remember that these companies do not have much of a track-record in terms of capital return, everyone (founders, shareholders, consumers - everyone) may ultimately be at the mercy of the government, technology is difficult to predict, the local currency might depreciate versus the dollar (a lot), international investors may not truly know what is going on on the ground in China, the legal structure is … suboptimal, and founders / managers are far from perfect.   I am inclined to believe that a good time to go shopping in the sector is now – in a negative sentiment phase, as prices and valuations are moving down to levels that are more than reasonable, depending on what kind of discount you want for a China ADR VIE Internet share vs a non-China, non-ADR, non-VIE share (if that discount is so high that you believe life is too short to accept such binary risks – that is a valid way to go, and a matter of preference). Furthermore, when hunting for what to buy, I believe it is sensible to prioritize the players that have some sort of track record of withstanding competition or preserving and ideally growing market share, have some sort of analogy in the Western markets that is helpful in understanding the business model, and have proven themselves to be utilities to their customer bases.   Weibo satisfies these criteria. It trades at the lowest possible multiple of next 12 month consensus GAAP earnings in its history. On forward GAAP, it is the cheapest of the long-established listed China internet names. It has a formidable growth momentum and if this momentum continues propeling the business forward for just another 1-2 more years and the business is not viewed as a terminal decliner by then, the odds for a long here are rather favorable, in my view.

Weibo has been growing rapidly because it took advantage of a favourable market backdrop (digital ads; mobile; etc) but also innovated at a good pace.  Often described as the Twitter of China, in reality, it is much more than that; Weibo has products and functions that resemble elements of Facebook, Instagram and Reddit. It is a sprawling ecosystem that is undergoing a general tilt to video and still has significant upside versus peers when you focus on metrics such as ad pricing, ad load, revenue and EV per user or unit of time engagement.  The upside is such that there is plenty of room to haircut Weibo’s metrics and still get to a conclusion that Weibo’s growth momentum, should it continue at least 1-2 more years, will carry it towards an EBIT not far off consensus numbers. I work off 2021 numbers. If Weibo delivers anything close to consensus estimates (2021 GAAP EPS of $5.45) and is then exited at a reasonable multiple of earnings (20x P/E; 16x net of cash), the stock price is set up for at least a double over 3 years.  In a downside case, if you see zero incremental EBIT growth after 2019 (i.e., get a 2021 GAAP EPS of $3.47) and then sell the business for 10x P/E ex- cumulative net cash, there is downside to a stock price of $43 (25% downside from current levels).


 
Some assorted thoughts / Q&A:

What are the consensus numbers, and what are they embedding?
On an LTM basis, Weibo is earning a GAAP EPS of $2.09; in the most recent quarter, it grew GAAP EPS by 188% yoy. Based on Weibo’s guidance, sell-side is expecting FY 2018 GAAP EPS of $2.50 (up 48% yoy) and is then further extrapolating the growth momentum (with some deceleration) for a FY 2019 GAAP EPS of $3.47 (up 38% yoy).  Sell-side is in the midst of a general China internet downgrade cycle, lowering their estimates ahead of Q3 reports that are coming out over the next 2.5 weeks. That is contributing to further share price declines; and as share prices drop, brokers have to lower their estimates again, exacerbating the stock price doom-loop. Reasons that the brokers cite for their incremental bearishness tend to centre around macro (advertising spend is not immune to cycles in the short- & medium term), with some added negativity around the potential impact of the government’s continued curtailing of the gaming and P2P sectors (but those make up less than 15% of Weibo’s revenue).   

Let’s examine that consensus FY 2019 GAAP EPS $3.47 figure, for a minute.  The current stock price implies a 16x GAAP P/E multiple on that number; if that consensus number is met, for the business to be fairly priced at that LTM multiple at exit (which also pencils out to less than that on a net-of cash basis), I think it would have to be viewed as a secularly-ex-growth business that is also headed into some sort of macro recession.

Why not invest via SINA?
Investing via SINA is a valid approach.  User ‘ahab931’ lays out the right (IMO) way to think about that choice in his write-up from 2017.  By my maths, at the date of that write-up, the implied NAV discount was 7% and today it is closer to 25%.  The highest it has ever been was 45% (in May of this year). China SOTP discounts are not new (e.g. see BABA/AABA/Yahoo, Naspers/Tencent, the SOHU complex).  Based on the mixed corporate governance history of SINA / Guowei Chao (e.g. the defeated activist initiative by Aristeia), some discount seems appropriate. At a 25% discount, I am somewhat indifferent between buying directly via WB or via SINA.  There are pros and cons in both. The pros of SINA include (i) SINA’s buyback authorization, (ii) SINA has more cash, securities and investments on that level, which could be valuable (e.g. the Didi stake from 2015 could be worth 3x+ cost), providing more balance-sheet support i.e. SINA trades at a 1.7x P/TBVPS, (iii) closer alignment with Guowei Chao, (iv) to the extent that SINA carries some costs or headcount on its books that are in reality associated with WB (i.e. overstating WB’s profitability), it makes more sense to invest via SINA.   The cons of investing via SINA (and thus, the pros of going directly into WB) include (i) SINA’s legacy operations are burning more and more cash (currently to the tune of $120m pa), as the SINA news/portal is apparently inferior relative to vibrant competition (BIDU; WeChat; Toutiao) and increasingly the only reason it gets any ads business is because the Weibo sales & marketing team bundles it when talking to clients; the SINA fintech business is also possibly cyclical and challenged by regulation, (ii) it’s not clear what appetite Guowei Chao has to exercise opex restraint and cut SINA-level headcount, (iii) Guowei Chao may not stop deploying cash to make more VC investments and I’m not well set-up to bet on his continued success with that, (iv) going via WB more closely aligns you with BABA, who has a 30% stake, (v) going via WB puts you closer to the actual value-driving assets, and (vi) these discounts can blow out if China-sentiment worsens.   These are qualitative and quantitative trade-offs. At current levels, I lean towards WB, but I respect either choice.

How has Weibo done in the context of the China ads market?
Weibo’s LTM revenue is up 75% yoy; revenue grew 75% also in 2017, and 37% in 2016.  These are rates that meaningfully exceed BIDU’s top-line growth for instance, which I consider the China Internet ads bellwether.  This outperformance is consistent with the ‘social’ segment of China’s digital ads market growing faster than search advertising. See chart below, from DB.   


This outperformance also was generally (at least, until recently) reflected by the relative stock chart.  



How is Weibo valued relative to comps? What is the monetization relative to comps?
Below is one way to think of Weibo’s valuation and profitability vs social media comps:


On a P/E basis, the LTM valuation is generally in-line; once you start looking out to FY 18E and 19E GAAP consensus, Weibo is meaningfully cheaper.  And then, once you start examining metrics such as EV/DAU or revenue/DAU, Weibo looks even cheaper. You would be correct in approaching these kinds of comparisons with some scepticism, because there are some differences in how the reported MAU and DAU metrics are calculated by these platforms (and when one of either DAU or MAU is not reported, they are estimated using engagement ratios).  In general, a Chinese social network’s MAU & DAU are overstated because (i) they include instances where a user logs into a 3rd-party platform or service using their Weibo or Wechat credentials, and (ii) anecdotally, the bot problem is a bigger issue for Chinese social networks than it is for Facebook, for instance. You can therefore pick a % haircut and apply it to the numbers, e.g. if you think that Weibo’s “true” DAU is 25% lower than the reported figure, the EV/DAU and revenue/DAU figure is 33% higher.   A second pushback against comping Weibo’s EV/DAU and rev/DAU vs Facebook or Twitter is that we are comparing different geographical markets, with the associated differences in affluence levels. This critique can be countered by looking at Facebook’s “Asia” numbers - Facebook Asia generates rev/DAU of $17 (vs $35 for Facebook consolidated). We can then go further to control for more granular levels of engagement, i.e. # of minutes spent per day on the platform; here, I use the best desktop-research estimates that I could find.   These numbers differ slightly from the prior table because the prior table uses LTM figures, whereas here I annualize the last reported quarter.




These stats are consistent with differences in the ad-load estimates (4.5% ad load rate for Weibo vs 12% Facebook Mobile, 7.5% Snapchat, 12% Toutiao).  In forecasting continued top-line growth of these platforms, the market is implicitly assuming that all these metrics (DAU and rev per DAU) will grow. I modelled these items out myself, using assumptions that net out at 2021 EPS 25% below consensus estimates (still enough to get a 15% IRR), and these assumptions correspond to DAU and rev/DAU vs street consensus the following way:


How is Weibo’s engagement doing relative to competition?
Generally speaking, there are 5 main digital internet ecosystems in China: Baidu, Tencent, Weibo/Sina, Alibaba, and Bytedance (i.e. Toutiao and Douyin/Tiktok); these ecosystems compete for ad dollars on one side, user eyeballs on the other, and the loyalty of celebrities/super-users/KOLs (key opinion leaders) on yet another side.  This essentially makes these platforms function like 3-sided networks. On all 5 ecosystems, the shift to mobile is essentially complete; e.g. 90%+ of Weibo’s DAUs are mobile, and 80%+ of Weibo’s revenue is mobile.

Thus far, Weibo’s user base and engagement are doing okay, with the DAU base growing by mid-teens % annualized, and user minutes also growing.  The big share gainer in the market in the past 12 months is Bytedance (via Douyin/TikTok and Toutiao), while the share donor is Tencent.



What are the recent developments in the social market in general, and how is Weibo responding to them?
There have been four broad themes on the user and product front in the China social internet market.  

The first theme is tightening regulation from the government on who can use the platforms and what can be said on them.  For instance, this week there is news of stricter age controls for one of Tencent’s games. Last year, the government begun requiring that new Weibo users provide their real names.  Earlier this month, the government required Weibo to bar children younger than 14 years of age from participating on the platform (the impact of this is not fatal, as users under 18 years of age make up less than 15% of the user base).   Overall, Weibo is notorious for being tightly censored; Guowei Chao is rumoured to have a good relationship with the government and also tends to keep a low profile (which I believe is a positive).

The second theme is a shift to video on all fronts (i.e. both the format of ads served, type of content that individual users prefer to consume, and the type of content they prefer to share – video is rising throughout).  Within this, the fastest growing user-generated format is short videos, as pioneered by Bytedance’s Douyin platform; that segment quickly became crowded, with each of the 4 other ecosystem leaders (Weibo included) launching their own versions of short video or incorporating elements of Douyin’s functionality (e.g. vertical format); this has resulted in Douyin’s growth flattening in recent months.

The third theme is the increasing importance of KOL (key opinion leader) advertising.  KOLs are essentially celebrity or quasi-celebrity bloggers that accept payment from advertisers in exchange for promotion, and can make more than tens of thousands of dollars a month from Weibo.  In the crowded and highly-sophisticated Chinese digital space, brands and advertisers can find that it is hard for them to get the consumers attention through classical advertising content and principles, and find a better ROI by relying on getting promoted by recognizable KOLs that are trusted, engaging and command the attention of large audiences.  Weibo appears to have been an early proponent of this trend (it is even possible that they created it). In 2014, Weibo divided its platform into 50+ verticals (e.g. beauty; apparel; gaming, etc), and seeded ‘power users’ within each category who eventually became quasi-celebrities in their own right, i.e. KOLs. These KOLs attract hundreds of thousands of followers, and their cachet has in turn attracted advertisers.  Many KOLs themselves are now also powerful Taobao sellers. The KOL ecosystem is now a distinguishing feature of Weibo, and Weibo is effectively acting as a platform/agent for the KOLs, managing the interactions between advertisers and KOLs, and collecting large fees in return. Weibo’s sophistication in this mirrors Weibo’s reputation as the most advanced platform in terms of targeting, i.e. the granularity of the ad campaigns that Weibo can run for the advertiser is understood to be unmatched in China (at least, for the time being).   KOLs are generally on multiple platforms, with significant focus on the top platforms such as Weibo and WeChat. The chart below illustrates Weibo’s power as a KOL advertising platform relative to some alternatives such as Douyin and WeChat, with a significantly higher reach. If we control for engagement, e.g. if a post on Weibo gets 1/3rd of the engagement intensity of a post on WeChat (the current market assumption), a promoted Weibo post is still a better investment, with 1,500,000 engaged impressions vs 105,000 engaged impressions for WeChat.  I ran various other calculations e.g. even if Weibo’s fee rate was doubled, its advertising efficiency would be still higher than on other platforms.


The fourth general theme is intensifying competition in a maturing (but likely not yet mature) market, with the players having to increasingly rely on competitive time-share and ad dollar-share gains to continue delivering fast growth.  Here, Tencent and Alibaba are particularly important. Tencent is altering Wechat with the aim of making it more attractive for advertisers and KOLs through mini-apps. Alibaba is a special case. In many ways, Alibaba already owns a direct relationship with the customer (i.e. over 50% of Chinese internet shoppers begin their search on Tmall or Taobao instead of a search engine such as Baidu, a significantly higher percentage than in markets such as the US).  However, Alibaba is also so big that it can and does spend on advertising on all non-Alibaba properties i.e, including Tencent, Baidu, Weibo, Douyin, etc. Alibaba currently accounts for 7% of Weibo’s revenue, making it Weibo’s largest single advertiser; Alibaba and Weibo’s cooperation was governed by a contract that expired 2 years ago; Weibo’s Alibaba revenue derived in the quarters after that expiry, but has since recovered and is growing again. Alibaba has also for years been trying to increase its control over the customer even further by pushing its own apps such as Weitao and Lu Ke.  Thus far, I understand those efforts have had limited market traction and have not had a particularly noticeable impact on Weibo’s engagement. It is likely that Alibaba will continue spreading its marketing dollars widely. Alibaba’s size and importance to the Chinese digital ecosystem means that Weibo’s fortunes will to some extent depend on Alibaba. But Alibaba’s 30% stake in Weibo also means that Alibaba has an interest in Weibo’s continued success (and Alibaba’s CEO Daniel Zhang sits on Weibo’s board). A sell-side analyst I spoke to believes that the stake could eventually lead to a full acquisition of Sina/Weibo by Alibaba.  Sina’s market cap currently is $4.4 billion. It is probably safe to say that acquiring it (and thereby getting control of Weibo) would not strain BABA’s finances, at least relative to the strategic value of the asset. It is unclear what price would be acceptable for Guowei Chao, but he has been described to me as a ‘financially-focused’ individual.

What kind of operational leverage is there in this business?
Gross margins are at 84% and are expanding by 3-5 pp each year, though this expected to flatten out at 85%.  R&D costs are now at 14% of sales, down from 25-30% 2 years ago. G&A costs are at 2.5% of sales, down from 6% 2 years ago.   EBIT margin is at 34% and has been relatively flat for the last 2 quarters. This is because the only cost item that is meaningfully growing is sales & marketing, at 33% of sales in the most recent quarter vs 24% last year.  This is apparently due to the increasing cost of adding incremental users in smaller cities and via pre-install fees paid to handset makers, as well as the increasing cost of servicing SME advertisers (though this headwind should moderate as Weibo is moving an increasing share of SME advertisers to programmatic).  I have the sales & marketing cost intensity declining to 29% of revenue in 2021, with an 2021 EBIT margin of 41%. In doing this, I am effectively assuming that Weibo’s sales & marketing efficiency (which dipped in the past 2 quarters) improves closer to Weibo’s historical means; see table below, where I am making some assumptions on DAU and advertiser churn.  

Conclusion
Weibo has attributes that we typically associate with high-quality businesses: high ROCE, cleanliness of the balance sheet, margin profile, cross-side network effects and same-side networks effects.  In a decade where various China consumer internet phenomena appeared, rapidly grew and then flatlined or declined, Weibo is one of those that has reached escape velocity and carved out a role as a ‘utility’ service for users as well as KOLs; the size and diversity of its social advertising ecosystem is in some ways comparable to BIDU, Tencent and BABA, but it is forgotten and potentially written off as ‘the Twitter of China’ (and even the negative connotation of that reference is becoming outdated as Twitter appears to be turning itself around; maybe the clown car fell into the gold mine… but it was still a gold mine).  China internet is a dynamic, fast-moving space; whatever the future of China internet looks like 3 years from now, I believe that Weibo’s size, diversity and current growth momentum will enable it to (i) position itself appropriately to take advantage of that future, and (ii) continue deriving a significant share of the digital advertising pie (and, even in the face of macro uncertainty, I expect China’s social advertising budget to be greater in 3-5 years from now than it is today). The price chart suggests that any growth momentum is gone (across the entire sector), but I have not found much hard evidence yet that it is truly gone and will not return again. Even if Weibo does not grow as fast in the future than it is doing today, this is a fundamentally strategic asset that is undervalued on user metrics.  I believe the stock is a double over 3 years.

Catalyst

 - Quarterly results and guidance meeting analyst expectations

 - Bottoming of the China-sentiment resulting in a multiple expansion for China internet stocks