Facebook inc the initial public offering case study

W12453

FACEBOOK, INC: THE INITIAL PUBLIC OFFERING (A)1 Ken Mark wrote this case under the supervision of Professors Deborah Compeau, Craig Dunbar and Michael R. King solely to provide material for class discussion. The authors do not intend to illustrate either effective or ineffective handling of a managerial situation. The authors may have disguised certain names and other identifying information to protect confidentiality. This publication may not be transmitted, photocopied, digitized or otherwise reproduced in any form or by any means without the permission of the copyright holder. Reproduction of this material is not covered under authorization by any reproduction rights organization. To order copies or request permission to reproduce materials, contact Ivey Publishing, Ivey Business School, Western University, London, Ontario, Canada, N6G 0N1; (t) 519.661.3208; (e) cases@ivey.ca; www.iveycases.com. Copyright © 2012, Richard Ivey School of Business Foundation Version: 2014-03-13 INTRODUCTION “The entire market is waiting for the emergence of Facebook as a publicly traded company,” said Jonathan McNeil, lead analyst at CXTechnology Fund (CXT), as he spoke to the fund’s investment committee on May 16, 2012. The highly anticipated pricing of the Facebook initial public offering (IPO) was underway, and in three hours, McNeil was scheduled to provide the lead underwriter, Morgan Stanley, with CXT’s final indication of his interest in the deal. Gesturing to Facebook’s preliminary prospectus (“Red Herring”), McNeil continued, “We have done our analysis, and we would like to present our recommendation on whether or not to buy shares in Facebook’s IPO.” Having been marketed with an initial price range in the high $20s to mid-$30s per share, the price talk for Facebook’s IPO had been increased to $34 to $38, valuing the eight-year-old company at over $100 billion. This price would make it the largest IPO of the year and the second largest IPO in U.S. history. The deal appeared to be oversubscribed with heavy interest from institutional and retail investors alike. But the valuation — at nearly 100 times trailing 12-month earnings and 26 times trailing 12-month sales — seemed expensive, even by technology standards. Yet, Facebook had changed the way consumers interacted online, spearheading the rise of social media. This explosive growth seemed poised to alter the way firms spent their advertising dollars, and Facebook was well-positioned to capture a growing share. COMPANY HISTORY AND OVERVIEW

Facebook was not originally created to be a company. It was built to accomplish a social mission — to make the world more open and connected.

Mark Zuckerberg, Facebook preliminary prospectus, May 16, 2012 Facebook was launched in February 2004 by Mark Zuckerberg and four roommates at Harvard University. The site was named after the popular directories circulated by different Harvard residences that featured a student’s picture beside his or her face. Facebook was designed as a social utility to allow friends to

1 This case has been written on the basis of published sources only. Consequently, the interpretation and perspective presented in this case are not necessarily those of Facebook or any of its employees.

This document is authorized for use only by kelly walsh (kellyannw92@gmail.com). Copying or posting is an infringement of copyright. Please contact customerservice@harvardbusiness.org or 800-988-0886 for additional copies.

mailto:cases@ivey.ca
http://www.iveycases.com/
Page 2 9B12N031

connect with each other over the Internet. After an initial run-in with the university administration, the Harvard site took off, leading Zuckerberg to expand to other U.S. and Canadian universities. By mid-2004, Zuckerberg had dropped out of Harvard, incorporated Facebook and moved operations to Palo Alto, California, where the company attracted its first investor, PayPal co-founder Peter Thiel. By year-end 2006, Facebook was open to anyone over 13 years old, had attracted an estimated 12 million users and was the seventh most heavily trafficked site on the Internet. In March 2006, Zuckerberg declined an offer to sell the company for $750 million, arguing it was worth $2 billion.2 His optimism was confirmed in October 2007 when Microsoft bought a 1.6 per cent stake for $240 million, valuing Facebook at $15 billion.3 Facebook continued its rapid growth, doubling its active users to 200 million between August 2008 and April 2009.4 To help manage the firm’s growth, Zuckerberg brought in seasoned executives Sheryl Sandberg as chief operating officer and David Ebersman as chief financial officer. In September 2009, Zuckerberg blogged that Facebook had reached 300 million users and was cash flow positive. Facebook’s users continued to grow at an extraordinary pace, passing 500 million users by July 2010, 800 million by September 2011 and 900 million by April 2012. Exhibit 1 provides a timeline that tracks Facebook’s growth. Over this period, Facebook had raised capital from angel investors such as Mark Andreessen, Reid Hoffman and Mark Pincus, and venture capitalists such as Accel Partners, Greylock Partners and Meritech Capital Partners. Based on transactions reported on SecondMarket Inc. and SharesPost — both online platforms for trading shares privately pre-IPO — Facebook’s implied value in December 2010 was between $41 billion to $57 billion, triple the amount since the Microsoft investment.5 Given the rising popularity and visibility of social media companies, financial market participants knew it was only a matter of time before Facebook went public. The initial Red Herring circulated by the underwriters in February 2012 announced Facebook’s plans to sell an unspecified amount of Class A common stock. The principal purposes of the IPO were to create a public market for the existing shareholders and to enable future access to the public equity markets. The proceeds would be used for working capital and other general corporate purposes. FACEBOOK’S BUSINESS MODEL Facebook provided an Internet platform that allowed its users to share comments, upload photos and recommend experiences (likes) to friends and family. Citing an industry report from August 2011, Facebook’s prospectus boldly stated that its goal was to connect all two billion global Internet users. For the fiscal year ending December 31, 2011, Facebook generated $1 billion in net income on total revenues of $3.7 billion, an increase of 65 per cent and 88 per cent respectively from a year earlier. Exhibit 2 provides Facebook’s consolidated financial statements. Advertising accounted for 98 per cent of Facebook’s revenues in 2009, 95 per cent in 2010 and 85 per cent in 2011. Facebook offered advertisers the opportunity to segment and target its users based on their demographic information, expressed interests and social connections. Facebook required users to disclose their authentic identity online. Any information uploaded to Facebook became the property of the firm.

2 http://www.businessweek.com/stories/2006-03-27/facebooks-on-the-block, accessed October 20, 2012. 3 http://blog.facebook.com/blog.php?post=72353897130, accessed October 20, 2012. 4 Ibid 5 http://www.bloomberg.com/news/2010-12-17/facebook-groupon-lead-54-rise-in-value-of-private-companies-report- find.html, accessed November 3, 2012.

This document is authorized for use only by kelly walsh (kellyannw92@gmail.com). Copying or posting is an infringement of copyright. Please contact customerservice@harvardbusiness.org or 800-988-0886 for additional copies.

http://www.businessweek.com/stories/2006-03-27/facebooks-on-the-block
http://www.bloomberg.com/news/2010-12-17/facebook-groupon-lead-54-rise-in-value-of-private-companies-report-find.html
http://www.bloomberg.com/news/2010-12-17/facebook-groupon-lead-54-rise-in-value-of-private-companies-report-find.html
Page 3 9B12N031

Facebook mapped the connections between users and their friends and recorded the products or services that they had “liked” in an extensive, proprietary database. Using this database, advertisers could target customized services and products based on users’ preferences and connections. Facebook called this feature “social context” and believed that advertising based on social context would be better received by consumers. Global advertising spending was estimated at $588 billion in 2011 and projected to reach $691 billion by 2015.6 Online advertising was projected to rise from $68 billion in 2010 to $120 billion in 2015. The balance of Facebook’s revenue was generated by its payments business, which came almost exclusively from the sale of virtual goods used in social games sold through the online gaming company, Zynga. Fees generated by these payments were $13 million in 2009, $106 million in 2010 and $557 million in 2011. In 2011, consumers purchased $9 billion worth of virtual goods from gaming and social networking sites and this market was forecast to grow to $14 billion by 2016. Facebook’s site was available in more than 70 different languages, and the company had offices or data centres in more than 20 countries. Geographically, about 56 per cent of Facebook’s 2011 revenues originated in the United States, down from 62 per cent in 2010. The majority of non-U.S. revenue came from Western Europe, Canada and Australia. MAUs, DAUs and ARPU Facebook categorized its users into monthly active users (MAUs), who visited the website in the last 30 days, and daily active users (DAUs), who were daily visitors. As of year-end 2011, Facebook reported 845 million MAUs, of which 161 million were based in the United States. While growth of U.S. MAUs was slowing, growth was picking up in emerging market economies such as Brazil and India. Facebook viewed DAUs and the ratio of DAUs to MAUs as a measure of user engagement. During December 2011, Facebook reported 483 million DAUs worldwide, an increase of 48 per cent versus a year earlier. DAUs as a percentage of MAUs increased from 54 per cent in December 2010 to 57 per cent in December 2011. Facebook also tracked users who accessed the site via a mobile app or mobile-optimized version of the website (mobile users). Increased mobile usage was a key contributor of growth with more than 425 million mobile MAUs in December 2011. Growth was driven by greater smartphone penetration in the United States and product enhancements across several mobile platforms. At the time of its IPO, Facebook could not display ads to mobile users. Increased use of this medium therefore threatened to cannibalize Facebook’s online advertising revenues unless it found a way around this obstacle. Facebook’s success in monetizing its customer base was measured by the average revenue per user (ARPU). Facebook defined ARPU as total revenue divided by the average of the MAUs at the beginning and the end of the year. Facebook’s ARPU was $5.11 in 2011. Exhibit 3 plots the growth of Facebook’s DAUs, MAUs, mobile MAUs and ARPUs over time. COMPETITIVE LANDSCAPE In the social networking space, Facebook competed on a global scale with MySpace, Google+, Twitter and LinkedIn. Facebook also faced stiff regional competition from Tencent, Renren and Sina Weibo in China;

6 http://www.sfgate.com/business/article/Global-Advertising-Industry-to-Reach-US-691-6-2455969.php, accessed October 12, 2012

This document is authorized for use only by kelly walsh (kellyannw92@gmail.com). Copying or posting is an infringement of copyright. Please contact customerservice@harvardbusiness.org or 800-988-0886 for additional copies.

http://www.sfgate.com/business/article/Global-Advertising-Industry-to-Reach-US-691-6-2455969.php
Page 4 9B12N031

mixi in Japan; Cyworld in Korea; Orkut (owned by Google) in Brazil and India; and vKontakte in Russia. Each company had a different business model and targeted specific customer segments. From 2005 until early 2008, MySpace had been the most visited social networking site in the world. The company was founded in late 2003 and bought by News Corporation less than two years later for US$580 million. In June 2006, MySpace had surpassed Google as the most visited website in the United States. By 2008, MySpace generated revenues of $800 million. Facebook overtook MySpace in the number of unique worldwide visitors in April 2008 and in the number of unique U.S. visitors in May 2009. The number of MySpace users had been declining steadily ever since. The lesson from MySpace’s rise and fall was not lost on McNeil, who had seen how easily a market leader could relinquish its lead. Google was started in early 1996 by two Stanford PhD students and went public in August 2004. Google had an advertising-based business model and generated almost all of its $38 billion in 2011 revenues from selling pay-per-click and site-specific advertising. With over 53,000 employees and a huge cash pile, Google could move rapidly. It had launched its own social networking service, Google+, in June 2011 and had already attracted 100 million active users by March 2012.7 Founded in 2006, Twitter’s microblogging service allowed users to send messages of up to 140 characters and had attracted over 500 million active users by year-end 2012.8 Twitter earned revenues from advertisers wanting to appear as part of a user’s Twitter feed.9 By December 2011 Twitter was valued at $8.4 billion although it remained privately owned. Twitter had forecast revenues of $110 million in 2011, up from $100 million in 2010.10 LinkedIn provided a social networking website for professionals that allowed them to post their employment history, then link their profile to other users with whom they had a professional connection. Founded in December 2002, LinkedIn had 175 million registered users by 2012, with revenues of $522 million and net income of $12 million.11 Users could access a basic version for free or pay $25 to $50 a month to access a premium version that allowed them to exchange messages and request introductions. Outside the social networking space, Facebook competed for advertisers’ dollars against leading online businesses such as Microsoft, Yahoo!, Amazon and eBay. ECONOMIC AND MARKET CONDITIONS Facebook’s IPO was moving forward during an improving — but still fragile — global economic environment. The world economy was still recovering from the 2007–09 global financial crisis, which had morphed by 2010 into a European sovereign debt crisis. The U.S. economy was slowly recovering with gross domestic product (GDP) forecast to grow by 2.2 per cent in 2012, up from 1.7 per cent in 2011, but still below the 3.3 per cent annual average from the 1980s and 1990s. U.S. unemployment remained stubbornly high above 8 per cent, while political partisanship in Washington ahead of the November 2012 presidential election threatened to derail the recovery. In particular, there were concerns that Democrats and Republicans would not be able to reach a consensus to fix the “fiscal cliff” — a series of tax and

7 http://google-plus.com/5746/google-crosses-100-million-active-users-in-march-2012-according-to-larry-page/, accessed October 20, 2012. 8 http://www.mediabistro.com/alltwitter/500-million-registered-users_b18842, accessed October 20, 2012. 9 http://news.cnet.com/8301-1023_3-57394477-93/the-$1-per-month-twitter-business-model/, accessed October 20, 2012. 10 http://online.wsj.com/article/SB10001424052748703716904576134543029279426.html?KEYWORDS=twitter, accessed November 7, 2012. 11 http://press.linkedin.com/about, accessed November 7, 2012.

This document is authorized for use only by kelly walsh (kellyannw92@gmail.com). Copying or posting is an infringement of copyright. Please contact customerservice@harvardbusiness.org or 800-988-0886 for additional copies.

http://google-plus.com/5746/google-crosses-100-million-active-users-in-march-2012-according-to-larry-page/
http://www.mediabistro.com/alltwitter/500-million-registered-users_b18842
http://news.cnet.com/8301-1023_3-57394477-93/the-$1-per-month-twitter-business-model/
http://online.wsj.com/article/SB10001424052748703716904576134543029279426.html?KEYWORDS=twitter
http://press.linkedin.com/about
Page 5 9B12N031

spending cuts that would automatically take effect at year-end. The picture abroad looked no better with Europe falling back into a recession while the powerhouse emerging market economies of China, Brazil and India showed signs of faltering. The U.S. stock markets had seen a strong run-up over the year to May 2012, with the S&P 500 Index rising 21 per cent from its lows in November 2011. Faced with the deteriorating economic outlook and political gridlock in the United States and Europe, investors had turned bearish, with the S&P 500 Index falling by 5 per cent in the first half of May. The tech-heavy NASDAQ 100 Index rose 17 per cent from mid- December 2011 to mid-May 2012 but seemed to have stalled recently. Exhibit 4 charts the recent performance of the NASDAQ 100 Index, the S&P 500 Index and the Internet Software & Services segment. The market volatility and continuing economic uncertainty had left the global IPO markets in the doldrums. During the first quarter of 2012, global IPO activity fell to $14.3 billion, down significantly from $46.6 billion during the first quarter of 2011. Exhibit 5 charts the number of IPOs from 2004 to 2012. McNeil and his team had carefully analyzed the performance of recent IPOs by LinkedIn, Groupon and Zynga (see Exhibit 6). In May 2011, LinkedIn had issued 7.84 million shares at $45 each for gross proceeds of $353 million, valuing the firm at $4.3 billion.12 Due to the popularity of the deal, LinkedIn had increased its price talk from a range of $32 to $35 to a range of $42 to $45 on the day before the pricing.13 Despite pricing the deal at the high end of the range, LinkedIn’s shares rose by 109 per cent on the first day of trading to close at $94.25. LinkedIn’s shares rose over the next year to $110.56 for a total gain of 146 per cent. The “deal-of-the-day” coupon company Groupon went public in November 2011, raising $700 million in the largest U.S. tech IPO since Google. Due to strong investor demand, Groupon’s underwriters had increased the number of shares offered from 30 million to 35 million and had priced the shares at $20, above the initial range of $16 to $18.14 This price valued the three-year-old company at $12.7 billion.15 Groupon’s shares rose 43 per cent on its first day of trading. After one week, its shares were still up by 21.3 per cent, but by mid-May its shares had fallen to $12.17, a loss of about 39 per cent post-IPO. Finally, the online gaming company Zynga went public in December 2011, selling 100 million shares at $10.00 per share. The deal was priced at the high end of the price talk of $8.50 to $10.00 and valued the four-year-old company at $7 billion.16 Zynga’s share price fell by 5 per cent on the first day of trading, and by mid-May its shares were trading at $8.56, 14.4 per cent below the IPO price. OTHER DEAL TERMS McNeil and his team pored over Facebook’s Red Herring to gain vital information about the offering (see Exhibit 7). A number of items caught their attention.

12 http://blogs.computerworld.com/18311/linkedin_ipo_stock_price_45_valuation_4_3b_date_5_19_symbol_lnkd, accessed November 7, 2012. 13 http://socialtimes.com/linkedin-ipo-7-84m-shares-at-32-35-each_b61483, accessed October 20, 2012. 14 http://www.reuters.com/article/2011/11/04/us-groupon-idUSTRE7A352020111104, accessed October 20, 2012. 15 http://digital-stats.blogspot.ca/2011/11/groupons-ipo-values-company-at-1265bn.html, accessed November 7, 2012. 16 http://money.cnn.com/2011/12/14/technology/zynga_ipo_price/index.htm, accessed November 7, 2012.

This document is authorized for use only by kelly walsh (kellyannw92@gmail.com). Copying or posting is an infringement of copyright. Please contact customerservice@harvardbusiness.org or 800-988-0886 for additional copies.

http://blogs.computerworld.com/18311/linkedin_ipo_stock_price_45_valuation_4_3b_date_5_19_symbol_lnkd
http://socialtimes.com/linkedin-ipo-7-84m-shares-at-32-35-each_b61483
http://www.reuters.com/article/2011/11/04/us-groupon-idUSTRE7A352020111104
http://digital-stats.blogspot.ca/2011/11/groupons-ipo-values-company-at-1265bn.html
http://money.cnn.com/2011/12/14/technology/zynga_ipo_price/index.htm
Page 6 9B12N031

Sales By Current Shareholders The Red Herring dated May 15 stated that Facebook was planning to sell 421,233,615 shares of Class A common stock. Of this amount, Facebook was issuing 180,000,000 shares with the remaining 241,233,615 shares sold by existing stockholders. As a result, Facebook would raise $6.1 billion to $6.8 billion while insiders would receive $8.1 billion to $9.1 billion. While McNeil knew an IPO was the moment for venture capitalists to take some money off the table, the sales by Zuckerberg and other insiders had to be taken into consideration. Exhibit 8 provides a list of shareholders and how many shares each was selling in the IPO (not including shares to be sold if the underwriters’ option was exercised in full). McNeil noted that there were five “lock-up” periods specifying when insiders could sell additional shares, ranging from 91 days to 366 days after the IPO. These lock-ups affected a total of 1.872 billion shares out of the 2.138 billion that would be outstanding post-IPO (see Exhibit 9). Dual-Class Share Structure Facebook had two classes of common shares, Class A and Class B, which had the same claim on the firm’s earnings but different voting rights. Each Class A share was entitled to one vote while a Class B share was entitled to 10 votes. Not surprisingly, the Class A shares were being sold in the IPO while the Class B shares were held exclusively by Facebook insiders and would remain unlisted. Assuming that 180,000,000 new Class A shares were issued in the IPO, Facebook would have 635,881,796 Class A shares and 1,502,203,241 Class B shares outstanding, with Class A shareholders controlling 4 per cent of the votes and Class B shareholders controlling the remainder. Through his ownership of Class B shares, Zuckerberg would directly and indirectly control 56 per cent of the votes. The Red Herring explained what this meant:

Mr. Zuckerberg has the ability to control the outcome of matters submitted to Facebook’s stockholders for approval, including the election of directors and any merger, consolidation, or sale of all or substantially all of our assets. This concentrated control could delay, defer, or prevent a change of control, merger, consolidation, or sale of all or substantially all of our assets that other stockholders support, or conversely this concentrated control could result in the consummation of such a transaction that other stockholders do not support.17

Zuckerberg had shown his willingness to use this control in the month prior to the IPO when he purchased Instagram — a popular online photo service — for $1 billion in cash and Facebook stock. Facebook’s board of directors had not been aware of the purchase until after the agreement had been reached.18 Fees Payable To The Underwriters Morgan Stanley was acting as lead underwriter for Facebook’s IPO, with J.P. Morgan and Goldman Sachs as joint leads, with 30 other co-managers. The lead underwriters managed the entire IPO process, from the preparation of the filing documents, organization of the roadshow, coordination of the book building, negotiation of the final pricing and distribution of the shares to their new owners. While the typical underwriting fee for an equity IPO was 3 per cent to 7 per cent of the amount being raised, Facebook would only pay 1.1 per cent reflecting both the size of the IPO and the prestige of Facebook.

17 “Facebook FORM S-1/A,” Red Herring, May 16, 2012, .p. 22. 18 http://www.informationweek.com/security/privacy/facebooks-history-from-dorm-to-ipo-darli/240000615?pgno=12, accessed October 20, 2012.

This document is authorized for use only by kelly walsh (kellyannw92@gmail.com). Copying or posting is an infringement of copyright. Please contact customerservice@harvardbusiness.org or 800-988-0886 for additional copies.

http://www.informationweek.com/security/privacy/facebooks-history-from-dorm-to-ipo-darli/240000615?pgno=12
Teletraan2
Highlight
Teletraan2
Highlight
Page 7 9B12N031

Underwriters had to manage numerous potential conflicts of interest in an IPO. They sought to build relationships with companies such as Facebook in the hopes of advising them on additional capital raisings or potential mergers and acquisitions. Facebook would want a successful IPO that raised as much capital as possible with the share price rising afterwards, setting the stage for future secondary offerings. The underwriters often had equity analysts who would initiate coverage of the company and issue price targets for the stock as well as an investment recommendation. The underwriters earned their fees by selling stock to their institutional and retail customers who wanted to buy the shares for as low a price as possible. Customers were particularly anxious to buy shares in “hot” IPOs where the shares were expected to “pop” on the first day by up to 35 per cent. In the event the issue did not “pop,” underwriters were expected to offer price support, which meant maintaining a floor price for the issue. The underwriters had an overallotment option (“greenshoe”) that allowed them to sell up to an additional 15 per cent of the offering. The underwriters could sell a total of 484 million shares even though they only had an allotment for 421 million. This greenshoe meant that the underwriters could effectively short 63 million shares. If the IPO was successful and the issue price rose beyond the offering price, the underwriters would exercise the greenshoe option with Facebook to cover their short position. If the issue was unsuccessful and the trading price threatened to fall below the IPO price, the underwriters would buy up shares in the market to cover their short position, providing price support for the issue. The underwriters would earn fees of 1.1 per cent on any shares sold in the IPO. FACEBOOK’S PRICE TALK Facebook had filed its first Red Herring on February 1, 2012, but the underwriters did not go out to investors with a formal price range until early May. At that time, the talk was in the range from the high $20s to mid-$30s per share. As momentum picked up and market conditions continued to improve, the underwriters launched the roadshow on May 7 with an eye to pricing the deal during the week of May 14. The amended preliminary prospectus filed on May 9 indicated that Facebook would sell 337,415,352 shares at a price between $28 and $35 per share. The amendment also indicated that the trend of Facebook’s DAU growth outpacing growth in the number of ads delivered had continued during the start of the second quarter of 2012. This trend was due to the increased usage of Facebook on mobile devices, in which display advertising was limited. As was customary, the preliminary prospectus contained no projections or other forward-looking information. The roadshow kicked-off with an investor presentation at the Sheraton Hotel in New York City featuring Zuckerberg, Ebersman and Sandberg. Led by Morgan Stanley, the road show included cross-country stops in cities where major institutional investors were located, including Boston, Chicago, Denver and Palo Alto. Facebook also released a YouTube video targeting retail investors. The roadshow wrapped up on Friday, May 11. The lead underwriters were actively soliciting investor interest through their sales teams. McNeil had been contacted by all three underwriters asking for his participation and interest. McNeil noted that the proposed price range was below the high of $44 per share that had been reached in March, based on a private deal posted on SharesPost.19 McNeil heard that there was significant institutional and retail demand for the deal, which he knew was only a preamble to the underwriters trying to raise the price range. At the same time, the lead managers seemed keen to keep the firm’s final IPO price conservative enough so that the shares could see a “pop” on the first day of trading. It was well documented that companies typically left money on the table, particularly when the price range was increased during the marketing of the IPO (see Exhibit 10).