We think businesses that possess unique and compounding datasets, which can be harnessed via artificial intelligence and machine learning principles to build new models that improve real-time customer or employee engagement, will be able to drive new sources of revenue and productivity in their business. The demand-side implications of AI and machine learning are very exciting to us - Curtis. Mohd Shahrin Yahya/The Edge
Click / Tap image to enlarge
The outlook for the global technology sector remains positive despite the recent rout in US tech stocks. In fact, it presents opportunities for investors to pick up high-quality counters at deep discounts, says Jonathan Curtis, vice-president of Franklin Equity Group at Franklin Templeton Investments.
“Investors with a short-term perspective tend to get nervous during these periods and sell their winners. This could be a huge mistake,” he adds.
The latest bout of underperformance (as well as the broader US market slump last month) comes amid strong US economic data in September. Reports of the unemployment rate falling to a 48-year low reignited market fears of accelerated US Federal Reserve interest rate hikes and how it would affect the tech sector going forward. This was compounded by lower-than-expected 3Q earnings calls, released throughout October.
That said, the tech sector’s fundamentals are expected to remain strong should any rate hikes occur. “On a relative basis, tech is the most cash-rich and among the most profitable sectors of the market right now. It does not need to tap the debt markets to operate in any way,” Curtis tells Personal Wealth.
The tech companies Franklin Templeton has invested in are largely independent of interest rate fluctuations. In fact, the firm tries to avoid investing in companies that are heavily dependent on cyclical factors, says Curtis.
Globally, tech stocks lost about US$1 trillion or 9% of their market value last month, according to Reuters. The underperformance is due to a number of reasons, including concerns about slowing global demand, stretched valuations and the ongoing trade tensions between the US and China.
The NYSE FANG+ Index shed more than 10% at end-July, down from its mid-June high of 3,062.88 points. The index tracks a group of highly traded growth stocks, including Facebook Inc, Apple Inc, Amazon.com Inc, Netflix Inc and Google’s parent, Alphabet Inc (FAANG). It also tracks Chinese tech giants Alibaba Group Holding Ltd, Baidu Inc and Tencent Holdings Ltd, as well as US-based electric-car maker Tesla and social media giant Twitter. Last month’s broad decline wiped US$172 billion off the value of the FAANG stocks.
Although the FAANG counters have been down 15% since June, they have outperformed the S&P 500 by about 6% year to date, says Curtis. “From a valuation perspective, tech is trading at just an 8% premium [over the wider index].”
Notable Chinese tech stocks have also suffered in other parts of the world, although it is mostly due to negative sentiment arising from the ongoing trade war between the US and China. “While it is causing a slowdown, it is not having a major impact [on our Chinese holdings]. Our two major holdings — Alibaba and Tencent — are not affected by much of the supply-chain activity that goes on in China,” says Curtis.
However, slowing growth in China could have an effect on Chinese tech stocks over the next 12 months, he adds. “We are definitely seeing some signs of a slowdown in China. Recent GDP figures slightly missed the 6.6% consensus. Instead, China’s GDP came in at 6.5%.”
Curtis says China’s multibillion-yuan stimulus package appears to be at the end of its cycle and this, too, could have an effect on Chinese tech counters. “We see a slight slowdown in this regard, although no major structural changes are expected to take place right now.”
The Chinese government has announced multiple rounds of tax cuts and infrastructure spending measures throughout the year in an effort to boost demand and compensate for its slowing economy.
There was yet another potential headwind for China’s tech companies when a Bloomberg report claimed last month that Chinese government agents had secretly inserted spy chips into the servers of small US-based chipmaker Super Micro, whose products are used by Amazon and Apple as well as sections of the US government and a host of other companies. This apparently gave China access to highly sensitive data.
Commenting on the allegations, Curtis says the focus of the report was on the building and manufacture of IT hardware in China. Therefore, it was unrelated to Franklin Templeton’s holdings in Alibaba and Tencent.
“Specific to the Bloomberg allegations, very reputable sources have disputed much of that story. However, I think it really highlights the global supply-chain risk, and this risk is not unique to China,” he says.
Upside potential on the demand side
Curtis says demand-side companies (those that provide goods and services to end users) have by far the most upside going forward. “We think businesses that possess unique and compounding datasets, which can be harnessed via artificial intelligence and machine learning principles to build new models that improve real-time customer or employee engagement, will be able to drive new sources of revenue and productivity in their business. The demand-side implications of AI and machine learning are very exciting to us.”
Salesforce.com Inc, Workday Inc and ServiceNow Inc are examples of demand-side companies that leverage AI and machine learning in their businesses. These cloud-based enterprise-software companies are strong performers and are in the top 10 holdings of the Franklin Technology Fund.
Salesforce.com specialises in customer relationship management products, sold using the software as a service (SaaS) model. Workday is an on-demand financial management and human capital management software vendor. ServiceNow is a provider of IT operations software to mid-sized and large corporations.
“These companies are using AI and machine learning to make their businesses stronger [in addition to their customers’]. As they are all SaaS companies, they see every bit of engagement on their platform. They track every click, every feature, and they get to see what the most popular and unpopular features are. This really drives productivity in their R&D process,” says Curtis.
Under the old vendor model, businesses purchased physical copies of enterprise software from local third-party vendors. “The vendor had no idea about the customer’s preferences or the features that saw the most use,” Curtis points out.
In addition to a comprehensive dataset on customer usage, these SaaS businesses also have access to their customers’ business data. “ServiceNow has all of its customers’ data on workflow. Salesforce.com has its clients’ customer data. WorkDay holds all of its customers’ data on employee records and company financials,” he says.
“All these companies are leveraging AI and machine learning to automate the process of sifting through all that data, acquiring unique insights from that data and selling those insights back to their customers. This is a stealth AI and machine learning play — one that is going to drive more sources of revenue and productivity for these businesses. That is why they are among our biggest overweights right now.”
Another under-the-radar tech play, according to Curtis, is Twilio. “I spoke about Twilio at The Edge-Citigold Wealth Forum 2018 in March and the stock has appreciated nearly 100% since. The company provides communication services to developers and businesses that want to engage with customers over the traditional telephone network in real-time, contextually relevant methods. It does this through voice and messaging applications. Every phone on the planet has these features, but not all phones have WhatsApp or WeChat,” he says.
The Franklin Technology Fund is the firm’s flagship technology fund. Launched in April 2000, it boasts US$3.08 billion in net assets and aims to provide capital appreciation by investing at least two-thirds of its assets in companies that are expected to benefit from the development, advancement and use of technology.
For the year to Sept 30, the fund was up 19.65%. It was up 26.12%, 24.78% and 18.75% over one, three and five years respectively.
Its top three holdings were Amazon, Microsoft Corp and Apple. Malaysian investors can gain access to this fund via the CIMB-Principal Global Technology Fund, which feeds into the Franklin Technology Fund.
Facebook still a strong play
One of the largest demand-side counters that Curtis is bullish on is social media giant Facebook, one-fifth of the much maligned FAANG stocks. He attributes some of its underperformance to company fundamentals.
“It has seen a bit of a slowdown in its core Facebook application. However, we are still seeing very strong Instagram growth. In addition, Facebook is investing heavily to ensure better safety on its platforms, all to give users confidence that theirs is a safe place in which to engage others,” he says.
Despite the company’s underperformance, Curtis believes Facebook has a lot of room for growth. The company currently has four options when it comes to revenue generation, but its performance is only driven by two of them.
First, the Facebook app has been a long-time earner for the company, even as it grapples with migrating audiences and flattening user engagement. “I think there have been concerns about the modestly slowing engagement on the core Facebook app, some of which is probably the result of reduced user confidence in the wake of many security or politically linked events,” says Curtis.
Facebook has come under heavy scrutiny in recent months over allegations of security breaches and user data being used to influence the 2016 US presidential election. In late September, the company revealed that its network had suffered a security attack that exposed the personal information of almost 50 million users. This was the largest breach in the company’s 14-year history, according to some news reports.
Cambridge Analytica, a political consulting firm that worked on the pro-Brexit campaign as well as US President Donald Trump’s 2016 election campaign, had collected the personal data of up to 87 million Facebook users. This resulted in a wave of negative press for Facebook and its shares plunged more than 24% during this period.
“Our belief is that Facebook has largely locked down a lot of those issues. There has been a massive increase in investment to make the platform safer and we actually like the investments that Facebook is making right now,” says Curtis.
“The first phase of investment — aimed at finding offensive content on the platform and removing it — is very labour-intensive. But over the long run, much of that labour will be augmented by Facebook’s very heavy investments in AI and machine learning.”
Instagram, Facebook’s 2012 acquisition and second revenue generator, has been a big win for the company. Since its US$1 billion purchase, the picture and video-sharing app has been a significant earner for the company. “Bear in mind that Instagram is still in the early days of monetisation,” says Curtis.
In an October article on tech news portal Recode, KeyBanc Capital Markets research analyst Andrew Hargreaves said Instagram generated roughly US$2 billion in the second quarter of this year alone. In the longer term, he expects Instagram to grow to about 30% of Facebook’s ad revenue over the next two years and to contribute nearly 70% of its new revenue by 2020.
Curtis sees a huge upside to Facebook’s other two apps — its Messenger app and the globally popular WhatsApp messaging service. Right now, Facebook’s performance is being driven by just half of its fully monetised assets as WhatsApp and Facebook Messenger services are provided at no charge.
“Facebook has started the process of monetising WhatsApp and Facebook Messenger. Both platforms have massive user communities, very light ad loads and very light monetisation of business-to-consumer engagement,” says Curtis.
“There is certainly a lot of business-to-consumer engagement taking place on both platforms. In the last one to two quarters, Facebook has been ramping up its monetisation activities on those fronts.”
The successful monetisation of these platforms could be a major game changer for Facebook. “You could build another Facebook just on the revenue that could be generated from monetising WhatsApp and Facebook Messenger activity,” he says.
Upside potential notwithstanding, Facebook’s stock has struggled this year amid security breaches, underwhelming 2Q earnings and political fallout. Curtis believes he knows why.
“We think investors can be incredibly short-sighted sometimes. They have not yet built into their models the incremental monetisation opportunities of WhatsApp and Facebook Messenger. Right now, Facebook is capturing no revenue from those platforms. But when it switches on those monetisation engines, we will start seeing very big surprises in its earnings reports. We think this will drive the stock even higher,” he says.
Under the circumstances, Curtis believes that the stock is comparatively inexpensive. “Our analysts like it because we think the long-term prospects are quite good. All businesses go through periods of investor scepticism, but the key is to take a multi-year outlook and consider what the business can look like at full maturity. I am telling you right now that Facebook Messenger and WhatsApp are not built into anybody’s models for the business right now and that is where we see big upside potential,” he says.
Back in December last year, Curtis predicted AI and machine learning would be the two major investment themes for this year. “These themes have fared very well for us. We are closely tracking the leading indicators for activity in this space,” he says.
“We have seen an increase in the number of AI and machine learning job openings and increased demand for the kinds of chips and semiconductors that drive AI platforms. Chipmakers and memory companies are doing very well right now.”
Curtis cites opportunities in supply-side tech companies. “There are many great businesses out there that sell enabling technologies such as chips, cloud services and storage infrastructure. These companies include KLA Tencor, Lam Research, Applied Materials, Nvidia, Samsung and Micron. We have holdings in these businesses.”
Despite the mixed 3Q performances in this subsector, he remains encouraged by the long-term prospects. “We really like the semiconductor capital equipment sector right now. We think it is trading at a deep discount relative to its intrinsic value. These companies sell key components to semi-conductor manufacturers. This is a sector that has been particularly beat down for a variety of short-term concerns,” he says.
“With Moore’s Law starting to slow down, capital intensity in the tech sector is on the rise. This is good news for equipment manufacturers. The demand for memory is on the rise because of the memory-intensive workloads inherent in AI and machine learning. This will require a lot of capacity and, ultimately, a lot of semiconductor capital equipment.”
Moore’s Law is the observation that the number of transistors on a chip doubles every year while costs are halved.
Curtis believes that there will be an acceleration of AI and machine learning-driven digital transformations next year. The wider digital transformation trend that is gathering pace among non-tech companies will create longer-term performance in the tech sector.
“We think businesses with very large and compounding datasets can be natural platform businesses with natural monopolies. This is simply by virtue of the data they collect and how they use that data to pull away and really prevent new competitors from entering their market,” he says.
“Google has a natural monopoly in search. Facebook has a natural monopoly in social media and advertising. Amazon has a natural monopoly largely because of its mastery of data, but also because of its e-commerce distribution platform.”
Meanwhile, Alibaba is building a natural e-commerce monopoly of its own. “If you listen to the start of any Alibaba engagement with its investors, it always starts by reminding investors that it is a data company. It will use data to not only deliver a tremendous experience for its end-customers but also its partners,” says Curtis.
In the current environment, he advises investors to adopt Franklin Templeton’s investing strategy during periods of volatility. “Take a long-term view, have a process to find secular winners and develop expertise in your investments. Get to know your investments very well,” he says.
“Do not invest in one-hit wonders. Instead, look for platform businesses with large, compounding datasets and seek out businesses that are building some kind of sustainable, unfair competitive advantage that will sustain them for years, not just through one economic or product cycle.”