Investor biases
A core belief of ours at Bireme Capital is that human cognitive biases drive security mispricings. This is not to say that we don’t believe in mostly-efficient markets, or that we are unaware that investor biases often cancel each other out. We simply believe that, occasionally, biased investors bunch together on a single side of the ledger, resulting in a mispriced stock.
Anchoring
“You’ll have to turn in your value investing card if you buy Facebook,” I was told at a recent meeting of value investors. The comment intrigued me. Is this a common perspective from value investors? How many haven’t analyzed this business because they assume it’s the type of stock Benjamin Graham would warn them about? The truth is that I once felt this way about Facebook. Whenever I saw the stock discussed on CNBC or held in a relative’s investment account, a little voice inside me scoffed. After all, Facebook is a high-flying growth stock -- definitely NOT a value stock -- and should therefore be shunned. This is a reaction that many value investors have evolved to have. It is protective against the plethora of overvalued stocks that litter the markets on any given day. In fact, historically there have been few better investing rules than to avoid the type of stocks that inspire bullish acronyms such as FANG or FADS CAN. While these stocks are exciting, fast-growing and well-known, all too often they are overvalued. Usually, the company’s vaunted prospects are already incorporated into the price. But if any investors still mentally group Facebook with the high-fliers, they are undoubtedly falling prey to anchoring bias, where individuals overly rely on initial pieces of information and fail to adequately update their opinions over time. With respect to Facebook, the anchor is likely to be years-old valuation metrics. As recently as 2016, FB traded for 80x earnings, exactly the type of multiple that value investors have learned to stay away from. But a lot has changed since then, and the company now trades for a reasonable ~22x trailing earnings. Value investors that ignore FB stock, such as the one who warned me that I’d lose my value investing card, have “anchored” FB in their minds as a hyper-growth, high-PE, likely-overvalued stock. They really should check the numbers. Availability The reason FB is no longer a high-PE stock is simple: earnings have grown faster than the stock price. One of the reasons the price has not kept pace is a recent plunge, from $220 in July 2018 to $130 at the end of the year. This allowed for our initial investment around $132. We think the severity of the decline was caused by availability bias. The availability heuristic is one of the most common cognitive biases, and causes humans to overweight more salient, or “available” information, and underweight all else. In the investing world, this can cause investors to weigh newsworthy events more heavily than the fundamentals of the underlying business. And there have been plenty of newsworthy events for Facebook investors to worry about. There has been a new scandal seemingly every month for the last few years. Wired has compiled a list of 21 scandals in the past couple of years, including:
While many of these stories sound scary, we don’t believe they’ve changed the fundamental trajectory of the business. Still, let’s look at how they might result in risks for Facebook. Users One way the negative publicity would hurt Facebook’s business is if users deserted the platform. There have been many news stories about this possibility (see here, here and here) since early 2018, many of them citing a Pew Research study saying that 44% of young people in the US have deleted the app in the past year. But what does the data say? First, it appears that the movement to #deleteFacebook was extremely short-lived. The frequency of searches for these two words had a small peak in early 2018, when the Cambridge Analytica scandal broke, but otherwise has been on a steady downward trend. It is currently at a nine year low. Second, third-party data shows that traffic to the site has been steady, with facebook.com consistently ranking as the third most popular website in the US according to Amazon-owned Alexa.com. In fact, it has been in the number three spot every single day since early 2018. Third, Facebook continues to grow users according to their own publicly disclosed figures. While the core Facebook site has reached saturation in the US and Europe, it has not lost material users in those countries. Elsewhere it continues to grow, expanding its user base 10-20% annually. The chart below shows users of the core Facebook app and facebook.com; it does not include Instagram or WhatsApp. I believe that an analysis of Instagram user numbers—which Facebook does not generally disclose—would show steady growth and no effect from the latest Facebook-related scandals. As I noted in our Q4 letter, 57% of Americans aren’t even aware of the link between Facebook and Instagram. And more than 80% of Instagram’s users are outside the US, where Facebook’s scandals are likely to be much less prominent. While the decline in Facebook’s stock price may indicate investors are concerned the company will lose users, we prefer to trust the data. It shows little impact to date. Regulatory risk Increased regulation is another obvious risk stemming from the scandals above, and many believe a crackdown on Facebook is imminent (here, here, and here). The investment community is taking this risk seriously as well, as evidenced by the 7% fall in the stock on December 19th, when it was revealed that the Attorney General of Washington, DC is suing the company. While future regulations may increase costs, they shouldn't dramatically alter the long-term growth trajectory of the business. One reason I believe this is the relatively small impact of recent European regulation. Commonly shortened to “GDPR,” the General Data Protection Regulation is one of the most sweeping privacy and data security regulations the world has seen. The rules require, among other things:
The regulation went live in Q2 2018, but the general consensus is that the law has, if anything, strengthened the hold of Google and Facebook on the European digital advertising market. These behemoths can put significant resources behind complying with the law while preserving ad targeting ability, while competing media firms simply don’t have the money or staff to pull this off. In fact, some have posited that the advantage Facebook has in navigating a larger regulatory burden may have inspired Zuckerberg’s recent request for increased government regulation of social media. The early data also show how little GDPR has impacted Facebook’s business directly. While Facebook user growth in Europe did slow in the second half of the past year, revenue per user continued to increase. European ARPU growth was 40%, 29%, and 24% in Q2-Q4 2018. These numbers were higher than the company's worldwide average. I think the European experience has shown that even if a stringent new law is enacted in the US, the business will probably be fine. And of course, other geographies would be totally unaffected. Margins FB has some of the highest margins of any company in the world -- nearly 50% in 2017. This compares favorably even to other internet giants like Google, Baidu, and Twitter. This makes sense for a few reasons. For one, FB doesn't have to pay traffic acquisition costs (the payments that Google and others make to partners to drive web traffic). Google spends 23% of advertising revenue on such payments. The scale of the business also helps. Facebook's 2+ billion users and high overall engagement mean that the somewhat fixed software development costs get spread over a wider base of revenues and users relative to a Snapchat or Twitter. However, the outlook for future margins is less clear. Zuckerberg has said that data privacy and content moderation costs -- as a direct result of the scandals -- will continue to increase even as revenue growth slows. Investors are particularly concerned with the company’s July guidance for a ~35% operating margin over the long term. However, I am inclined to believe that they are sandbagging a little bit, and expenses may not increase as fast as people expect. First of all, it doesn't really make sense that expenses would rise faster than revenue over the long term. That’s not to say they won’t rise in the short term: all of the initiatives that Facebook is undertaking in response to the recent problems imply a rise in expenses. But most fixes, such as improvements to data privacy protocols, ought to be one-time software patches or policy changes. Tighter moderation of abusive or questionable content can be performed primarily by software, and this software is improving all the time. My guess is that ten years from now, Facebook will need fewer, not more, of the human content moderators described in this Motherboard post. Facebook has also overestimated expense increases in the past. In 2016, 2017, and 2018, expense growth came in closer to the bottom end of Facebook’s guided range than the top. Our current expectation is that expenses will grow 42% in 2019, at the bottom end of Facebook’s range, and that growth in costs will slow materially after that. Thus, if we are correct, the responses to the recent scandals should result in only modest profit margin declines past 2019. We project terminal operating margins to be 38%, 3% higher than management’s guidance. Our projections While investors may be over-estimating the likely adverse effects of recent scandals, we think they are under-appreciating the company’s growth drivers going forward. Below is a discussion of some of the key assumptions that drive our five-year earnings projections, which forecast a rise in EPS from $7 to $12. This does not include any benefit from share buybacks. User growth in Asia and “Rest of World”: While we forecast flat users in the US and Europe, user growth in Asia and RoW is likely, and merely a continuation of a years-long trend. These regions are still relatively underpenetrated compared to North America and Europe. We expect total daily active users in all regions to grow from 1.5 billion to 1.9 billion over the next 5 years. ARPU growth across the board: Since going public in 2011, Facebook has disclosed quarterly ARPU across four major geographies. Of those 120 data points, only one quarter in one geography showed negative YoY ARPU growth. Some quarters ARPU grows 30%, some quarters it’s 10%. But it’s almost always positive in every geography. There are multiple drivers of this growth:
In terms of Facebook’s new platforms, Instagram is the crown jewel. It is not hyperbole to say it is one of the greatest acquisitions of all time. For a “mere” one billion dollars, Zuckerberg purchased an app that is now used by over a billion people, a figure which was growing over 30% annually the last time they released data publicly. The platform is a massive opportunity for Facebook, as Instagram is probably under-monetized. It is estimated that Instagram generates less than 20% of Facebook revenue. This is despite the fact that it likely accounts for more than one third of usage (excluding WhatsApp), over-indexes on developed countries, and consists of visual media, making the platform perfect for advertising. We are not alone in our high expectations for the site, with eMarketer projecting Instagram revenues to triple over the next three years: I think the ceiling for Instagram’s business may prove higher than the core Facebook app on a per-user basis, given the importance of the platform to consumer products brands. As an example of where it can go, look no further than the company’s new e-commerce feature, where users can browse, click, and pay all within the app. I think this is an enormous revenue opportunity and wouldn’t be surprised to see Instagram become a full-fledged competitor to Amazon in certain product categories. I expect Facebook to monetize the Instagram platform more effectively over time. When combined with Instagram’s likely user growth, the result will be a significant increase in Facebook’s revenue. Final conclusions Our overall view is that the scandals surrounding the company have dominated the news flow and caused investors, likely plagued by availability bias, to sell the stock down to unwarranted levels. Many value investors -- suffering from anchoring bias -- still assume the firm is priced as a high-growth stock, leaving a void of investor interest and an opportunity for outsized returns. We purchased the stock for our clients in the $130s. At that price, based on our free cash flow estimates and a 22x terminal multiple commensurate with the quality of the business, we estimated that Facebook would provide a return of roughly 15% annualized over the next five years. - Bireme Capital Appendix: Myspace risk This is an important question for the business, so while it didn’t fit into the overall scheme of post above, I thought it was worthwhile to share my thoughts on the topic Ever since Facebook went public, people have wondered, “Will Facebook one day go the way of Myspace or Friendster?” To demonstrate how prevalent this concern is, try googling "facebook myspace www.quora.com.” There are over 3,000 results, many of which ask this question. There are a multitude of reasons why Facebook has not yet become Myspace. But to better understand the long term risks to Facebook’s network, it is helpful to think about why it beat Myspace in the first place. Said another way: why did Myspace fail? The answer, of course, is everyone switched to Facebook. But why? You can read through the following links for some ideas on how Facebook pulled this off (source, source, source, source), but the posts boil down to the fact that Facebook simply executed better. However, a second key item is rarely discussed: Myspace’s lead was never that big anyway. When Facebook was started in February 2004, Myspace was still tiny. Over the next year Facebook remained confined to college campuses, growing slowly and by word of mouth only. In contrast, Myspace gained traction quickly, accelerating growth by using a large email database owned by parent company Intermix. When News Corp purchased the company for $580m in 2005, Myspace had about 20 million users. This meant that despite being four times as large as Facebook, they were still only used by a small fraction of the US population. The modest network effect Myspace enjoyed was therefore not insurmountable. A new social network, on the other hand, will face an extremely intense competitive environment. To gain traction, they must overcome Facebook and Instagram’s near-ubiquity and strong network effect. They must also overcome Facebook’s ability to copy any new features they might develop. Snapchat is, of course, the key cautionary tale in that vein. And even if a new social network were to gain hundreds of millions of users, it’s unclear if that would materially harm Facebook’s business. Look at the user figures for the core Facebook website (titled “Daily Active Facebook Users” above). Can you spot the point at which Instagram’s mushrooming 2013 and 2014 user numbers affected Facebook’s US or European user growth? I sure can’t. Instagram added more than 200 million users during this period. Bireme Capital LLC is a Registered Investment Advisor. Registration does not constitute an endorsement of the firm nor does it indicate that the advisor has attained a particular level of skill or ability. This piece is for informational purposes only. If not specified, quarter end values are used to calculate returns. While Bireme believes the sources of its information to be reliable, it makes no assurances to that effect. Bireme is also under no obligation to update this post should circumstances change. Nothing in this post should be construed as investment advice, and it is not an offer to sell or buy any security. Bireme clients may (and usually do) have positions in the securities mentioned. Comments are closed.
|