Please Note: Blog posts are not selected, edited or screened by Seeking Alpha editors.

2018 1H Performance:

|Includes: FB, T-Mobile US, Inc. (TMUS), WFC

Our gross return for the 1st half of 2018 is 4.46%, compared with 2.65% for S&P 500 in 2018. Since our inception 3 years ago, our annualized return is 16.8%, compared with 11.7% from S&P during the same period.





Full Year


S&P 500





(4.67% for 2nd half)


(-0.48% for 2nd half)


















(4.46% for 1H/2018)

2.65 %

During the first six month, Amazon and Facebook contribute majority of our gains. Financials (BRK, SYF, WFC) negative impact our results. The market is marked by extreme volatility in the 1st half on macro developments such as trade wars and higher inflation. We see several opportunity to deploy our excess cash. Throughout 1st half, we aggressively add our position in Facebook, Wells Fargo, and to a less extend in Broadcom. We also initiate a position in T-Mobile. Our current cash position is 10.0%.

Existing Holdings:


We more than doubled our position in FB during the wide ride of 1st half. Our timing is far from ideal. We were adding shares both before and during the Cambridge Analytica scandal. Our average price paid for FB is now $175. The share recover recently to $194 and FB is currently the 2nd largest position in our portfolio.

On the data scandal side, Cambridge Analytica marks one of the most difficult and transformative period of the company. The company has fundamentally changed its thinking on how to operate the platform and is now committed to solve a whole set of complex social issues. After CEO Mark Zuckerberg’s congress testimony, the company has roll out functions aiming at tracking political ads and shut down fake accounts. It also published guides for community standards. And we expect more tools coming for monitoring hate speeches. To be fair, there are still several ongoing investigations on Facebook. Such investigation or potential litigations can from time to time dramatically affect Facebook’s share price, thus affect our short term performance.

On the business side, while majority of attention is focused on the Facebook data scandal, a powerful secular trend is taking off underneath: both short and long mobile vertical video is gaining popularity around the world. And we believe Facebook will be one of largest beneficiary of this trend. For example in China, Douyin, a short format vertical video platform, has quickly amassed 150M DAU and 300 MAU less than two years of its launch. Its success has caught all three Chinese tech giants off the guard – particularly for Tencent. We think at least two reasons has restrained Tencent’s ability to duplicate Douyin success a large scale: powerful AI algorisms that enables high quality content feed, and massive data centers to support video steaming for a billion user.

FB does not has these two constraints. FB is the pioneer on AI content feed. In recent years, FB has beef up its data center CAPEX. Facebook’s CAPEX has gone up from $2.5B in 2015 to more than $14B in 2018 just to support its internal use. A scale rivals Google and Amazon, which are also public cloud providers. The bet is paying off as the company is now rolling out both short and long format video across its platforms. On Instagram and Whatapp, the DAU of Stories (short format vertical video) has grown exponentially to 400M and 450M, up almost 100% from last year. The average Stories post by user also dramatically increased as more functions, such as music background, are added into Stories.

More recently, Instagram launched IGTV, a dedicate app (also available inside IG) that support long format vertical videos up to one hour. Instagram is working directly with many influencers to produce creative contents that are 2-5 minutes long. In the back end, IG is providing influencers with state of art analytical tools. While it is very early, we think over time IGTV can be hugely popular and grab user time from YouTube. We are following the growth IGTV very closely and will report back to you in the year end.

Facebook’s valuation is also attractive not only compare to tech sector but also the broad market. Depend on memory pricing, we estimate that the company can generate from $35B to $40 free cash flow by 2020. At $194 and a market cap of $571 billion, the company is roughly trading around 14-16X FFCF. A multiple that is in line with the broad market but with a much faster growth rate. Facebook is also riding on several other secular trends such as AI, eCommerce, mobile payments, and VR/AR. In its most recent quarter, FB also another $9B to the buyback program, in addition to the $6B started in 2016. We think odds in our favor that Facebook will be much more valuable few years down the road. In the meantime , we will adjust our exposure depend on the development of investigations.

Wells Fargo:

Our thesis for adding Wells Fargo is straight forward. WFC’s core business remains strong, the share is cheap, and the company is repurchasing share at an very aggressive rate. We previously had a large position in WFC and has trimmed majority of our position after the Trump election. During 1st quarter, WFC was sold off on the news of Federal Reserve’s consent order on asset cap. (The company cannot grow its total asset over $1.95 trillion unless it gets regulators permission). The consent order is related to the prior issues which Wells Fargo has already made a lot of progress on. The company is now working hard to lift the asset cap by mid 2019, While the asset cap is a serious issue, the financial impact is minimal (less than $100M after tax impact).

More recently, Wells Fargo has passed Fed’s annual stress test. In the cycle from mid 2018 to mid 2019, the bank is allowed to repurchase up to $24.5B of its shares - $10B more than street consensus. The bank also increases its dividend 10% to $0.43/share. If WFC’s stock price remains unchanged and the company fully utilized its repurchase program, shareholder is on track for a 12% yield (9% buyback and 3% dividend).

Wells Fargo is now almost 3 years into its account fraud scandal. The bank has made countless changes to its operating structure, sold non-core businesses, and settle several key litigations. The core business of Wells Fargo remains one of the best franchises in corporate America. Most of its business lines are annuity businesses that generate strong cash flow and have best in class credit profile. The economic scale also allows WFC to spends billions on digital infrastructure, payment ecosystem and store upgrades, which overtime will greatly improve their franchise value compare to smaller competitors. While there is a lot number moving around ($2B cost savings, interest rate sensitivity , and run off of amortization for core deposits intangible), we estimate Wells Fargo can generate close to $30B net income by 2020. In meantime, Wells Fargo sits on more than $25B excess capital that are now beginning to return to shareholder over next 2-3 years. Such bargain is rare and we aggressively add to our position. Our average cost basis now $53.4. The stock ended the quarter slightly above our cost basis.


Our thesis for adding Broadcom is almost identical to adding WFC: strong fundamentals, cheap valuation, and large buybacks. Failing to acquire Qualcomm and uncertain demand on next generation iPhone are part of the reasons the stock has perform weakly the past 6 month. We added some position in Broadcom and our cost basis remain largely unchanged at $254. We have discussed Broadcom’s financial profile in previous letters and thus won’t repeat here again. In April, Broadcom authorized the repurchase by the Company of up to $12 billion of its common stock for the next 18 month. We are confident that this move will significant improve the shareholder value as the company is buying share at less than 12X free cash flow, while other analog competitors trades have much higher valuation. The company is has a dividend yield close to 3% and the dividend payout is almost for sure to go up in coming years.

New Holdings:


During 2Q/18, We initiate a position in TMUS. In April 29, TMUS announced it will to acquire Sprint through an all stock deal for $26B. Because most street analysts think the probabilities of the deal go through regulator is low, TMUS subsequently dropped from $64 to $56. Before the deal, TMUS is on track for a $9 billion, 3 year repurchase program. We saw an to build a position in a company that in our view will be much more valuable with or without the deal. Our average cost basis for TMUS is $59.0.

We have follow the telecom industry for a long time and has long admired John Legere remarkable skill at turning around company. Before joining TMUS, John Legere was the CEO of Global Crossing from 2001 to 2011. He led the company through bankruptcy and later sold it for $3B to Level 3 Communication (which we previously has invested). Since Joint TMUS in 2011, John Legere successfully implemented the famous “Un-carrier” program by focusing on what customer really want. From “un-contract” to “unlimited data” to “Netflix on us”, TMUS has dramatically the wireless landscape and its business prospers. From 2012-2017, TMUS has net added 13.8M post-paid phone customers. It total customers(including wholesale), has more doubled from 33.4M to 70.6M. Adjusted EBITDA went from $4.9B to $11.2B. Leveraging on its financial success, TMUS bid out competitors and spend $8B in 2017 to acquire a large amount of 600MHZ spectrum - tripling its low band spectrum capacity. TMUS is now gradually rolling out its 600MHZ LTE in cities that the company used to be weak on networks speed.

If T-Mobile is able to pull off the deal with Sprint, the combined business would have sufficient scale to build out an initial 5G network much earlier than competitor due to its ownership of both low-band (600 Mhz) and mid-band (2.5Ghz) spectrum. Large competitors (ATT & Verizon) have yet to figure a way to use mmWave to build out an economic variable 5G network though high band spectrum. More importantly, the combined business would have enough scale to take on large cable companies (Comcast and Charter), as more millennium are expected to only use wireless service in the 5G era. We think the upside is very high for TMUS if the deal go through, while the downside the limited because its business momentum.

Thoughts on Technology trend:

There are two distinct group of large technology companies: ones with extreme profitability and ones with no profitability in the foreseeable future.

  1. Ones with extreme profitability:

Clearly leading tech giants in this group. More remarkably, tech giants generate ever growing profit with very little tangible assets. Warren Buffett pointed out this trend in his 2018 Berkshire Hathaway meeting:

You now have four largest companies by market value that essentially need no net tangible asset. The earning of the S&P 500 as a percentage of their net tangible asset have gone up substantially during the last 20 years. And this has become somewhat can be called an asset-light economy. And I don’t think people has quite process all that”

To further illustrate, the following chart is a comparison of return on net tangible assets of the tech giants in their most recent fiscal year. Since all tech giants has net cash on hand that is not correlate to operate their core business, we exclude net cash in calculating their net tangible asset.






(Dallar/yuan @6.504)


(Dullar/yuan @6.504)

Revenue Growth Rate:







Operating Earning




(ex litigation)




(ex other gain)

Net Tangible Asset:







Return on Net Tangible Asset







As chart show above, tech giants shows a return on net tangible asset from well over 100% to “infinity”. Infinity is a result that the company has a negative tangible asset. Software companies such as Microsoft or Tencent which has large unearned revenue on their liability side usually has such characteristics. Simply put, they are running the company using other people’s money. We current have shares in four tech giants. We believe their risk reward profile continue to be attractive compare to most market alternatives. It’s not uncommon for mega-caps to get mispriced.

  1. Ones with no profitability in the foreseeable future.

On the other end of the spectrum, there are a range of large tech unicorns that do not generate earnings and may not generate earnings in the foreseeable future. Many unicorns, especially Chinese ones, are going public in the 2018 to 2020. Most of the unicorns are consumer brands. These include steaming companies like iQiYi(already public), O2O services companies like Meituan, share-economy companies like DiDi, Uber and Airbnb. Most of these companies are unprofitable at current level largely due to their large capital need for future growth. In addition to their core business, these companies also have capital need to penetrate into many adjacent industries. While we are familiar their services and admire their brand power, valuing them and applying an appropriate rate of margin of safety is a another question.

We do learned some from investing in companies with no profitability in past. We are lucky to hold on Amazon since 2015 and Amazon since has built up at three very profitable business: AWS, Amazon Prime and Retail 3rd Party services, In Cimpress, while we sold the share early on in 2016 for $86 at break-even price (vs $147 today), we learned a lot on how the owner compare IRR on investing in core business versus new ones. Since most unicorns will raise a large sum of capital through IPO, how they can intelligently allocate money in their core business versus new areas will be a key metric in determining their future value.

Growth companies with no profitability usually have very large price swings during their public journey. We may invest in a few of them from time to time. Rest assured, we will maintain a discipline of not chasing the momentum and only invest in the companies we know well and with enough margin of safety.


As we enter into the second half of the year, macro events continue to dominate headline: trade war officially starts, flattening yield curve, two more interest rate hike, Fed no longer funding the US government deficit when deficit goes up, China currency devaluation, A share in bear market, the list goes on. As usual, “preservation of capital” is the main purpose for this fund. Most of our purchase in 2018 following similar pattern: strong companies buying back their shares at below market multiples. While some names work out faster than others, we believe owning a group of them will work out in our favor over time. We look forward to report you back by the year end.