IQIYI: Blunting Wolfpack’s Fangs (NASDAQ:IQ)
My Chinese associates like to quote this saying “上得山多，终遇虎” which literally means those who go to the hills often would eventually ‘meet’ a tiger. In the context of COVID-19, if one chooses to ignore the warnings for social distancing and continue to gather in groups in close proximity, the person could eventually be infected by a spreader. In the context of investments, it’s a jungle out there and inevitably we encounter some ferocious wild animals who take a bite of our pockets or even eat us alive figuratively.
For us who choose to own the stocks of Chinese companies, there’s the perennial concern whether their financials are authentic and doubts fly whenever there are reports from short-sellers that appear highly convincing at first glance. To make the matter worse, once in a blue moon, we have some management having to admit to fraudulent practices in their companies.
Coincidentally or otherwise, we have the confluence of both short-seller reports and self-admissions within a week and three at that. Starbucks (SBUX) challenger in China Luckin Coffee (LK) was the first to drop a bombshell on April 2. A special committee brought to the attention of the company’s board of directors that COO Jian Liu and several employees reporting to him had “engaged in certain misconduct, including fabricating certain transactions” from the second quarter of 2019 to the fourth quarter of 2019 amounting to around RMB2.2 billion ($310 million).
Short-selling outfit Wolfpack Research locked its fangs on iQIYI
Hot on the heels of Luckin Coffee’s confession, famed short-seller Muddy Waters declared he’s short iQIYI (IQ) in a Tweet on Tuesday, claiming that the video-streaming company “fraudulently and materially overstates its users, revenues, acquisition consideration, and value of its ‘barter’ content.” The accusation was ostensibly based on a critical report prepared by Wolfpack Research with assistance from Muddy Waters.
Wolfpack estimated that iQIYI inflated its 2019 revenue by approximately RMB 8-13 billion, or 27 percent to 44 percent by overstating its user numbers by approximately 42 percent to 60 percent. Before exploring further, I checked out who’s behind Wolfpack and their past targets. Incidentally, they issued a report on news and video aggregator Qutoutiao (QTT) on December 10, 2019, exclaiming that 74 percent of its 2018 sales are ‘fake’.
Qutoutiao took more than two weeks to rebut Wolfpack but the stock had already begun to rebound days earlier. It subsequently continued on its uptrend and at the peak in February, the stock had more than doubled the price level when the report was released. Of course, the share price sunk since late-February but it’s very much due to the global sell down than anything the analysts from Wolfpack accused them of.
Looking further back, Wolfpack’s attack on Smart Global Holdings Inc (SGH) also did not see much success. Since the short report was released on September 25, 2019, the share price of Smart Global Holdings appreciated nearly 40 percent before slumping from the beginning of this year. Again, I doubt the decline was attributed to smoking guns exposed in Wolfpack’s report given the long time lag.
Anytime there is a short-seller report, even when the accusations could be refuted, there are probably some points raised that ought to be investigated further. Warren Buffett famously wrote in his annual letter to shareholders in Berkshire Hathaway (BRK.A)(BRK.B), “In the world of business, bad news often surfaces serially: you see a cockroach in your kitchen; as the days go by, you meet his relatives.”
In iQIYI’s case, I found the timing of Wolfpack’s report too opportunistic than a coincidence. It could even be half-baked, rushed out to take advantage of the Luckin Coffee fiasco and ongoing anti-Chinese sentiment amid the COVID-19 outbreak. Wolfpack’s track record doesn’t lend much credibility to its content. However, since veteran short-seller Carson Block of Muddy Waters was willing to stake his reputation on the thinly-staffed “research firm”, it warrants our attention.
I do not have the resources of these outfits but I will try to poke some holes in the report based on what I know to start the discussions. I will also share what my Chinese friends said about the debacle. This is not the first time I have endeavored to question short-seller reports.
I did it for Momo (MOMO) when Spruce Capital attacked the video-streaming provider for overpaying the Tantan acquisition, a discrepancy in its reported MAU growth and Momo’s reputation of being a (sex) cam site. Interested readers are welcome to check the article out for my thought-process regarding selected claims. Ben Axler, the founder of Spruce Capital, took the time to comment on the article.
Are iQIYI’s user numbers overstated?
Wolfpack fired off the series of accusations with the header “1. IQ Overstates its User Numbers” and the audacious claim that “our research uncovered data from three independent sources showing that IQ overstates its DAU numbers by 42% to 60%.” The researchers stated that two Chinese advertising companies provided them with data from iQIYI’s back-end system.
Whether Wolfpack indeed had access to the data, we wouldn’t know. However, the researchers admitted themselves that their projections for the DAU of September were based on “4 days of DAU data from the same week”. Is such sampling reasonable? What if the researchers took the worst four days of the month?
Wolfpack went on to calculate the mobile DAUs between what was reported by iQIYI and the ad agencies. To cut the story short, I reproduce the table in their report with the key figures.
Source: Wolfpack Research
There was no link to where the “175 million average mobile DAUs number disclosed by IQ in October 2019” came from. I did some Googling and discovered that Wang Xiaohui, the chief content officer of iQIYI, revealed (in Chinese) at a business development conference that the DAU from ‘independent devices’ on iQIYI’s platform was 175 million per day, with the monthly usage duration amounting to 7.7 billion hours. Now, since when are ‘independent’ devices the same as ‘mobile’ devices?
Seeking Alpha Premium member feanors highlighted the possibility that paid subscribers who are not shown advertisements are not included in the count of “eligible” DAU for the ad agencies, giving rise to a smaller figure than the total DAU. This seems logical as well. Furthermore, the 36 percent referring to the number of daily users is incorrect. It should instead be referring to the number of paid members in first-tier cities.
The researchers referenced the authoritative QuestMobile’s special report titled “China Mobile Internet Amid COVID-19 Plague” which stated that the average mobile DAUs were only 126.2 million during the first 10 days of the 2020 Chinese New Year. Again, the numbers cited by QuestMobile were for mobile devices, not all independent devices.
Furthermore, unlike TikTok, the leading short-video app, iQIYI’s contents are catered towards larger screens and have a much longer duration. This meant that a large portion of iQIYI’s DAU could be coming from desktops, laptops, and televisions. Hence, the mobile DAU as reported by QuestMobile would be just a fraction of iQIYI’s overall DAU.
The third supporting argument by Wolfpack for the DAU discrepancy was based on the “Heat Index” maps ironically created none other than iQIYI itself and made publicly available on its website. The researchers questioned why there should be a consistent pattern of provinces/zones ranked in the top 10 for most viewers” for areas with “low populations such as Macau, Hainan, Tibet, and Inner Mongolia.”
Source: Wolfpack Research
Seeking Alpha Premium member feanors made another excellent proposition that the index could be derived based on the popularity (per capita) of a province, rather than the absolute number of viewers. For instance, X number of users watching a particular show versus the total number of active users in the locale. Places with a much smaller population such as in Tibet and Hainan would then see the light of the day in such heat maps.
It wouldn’t be meaningful if the most populated provinces hog the leader board all the time. Therefore, I think what feanors said made great sense. In addition, I would think if iQIYI “employs methods to fraudulently inflate the viewership levels of its content” as alleged by Wolfpack, the executives would be smart enough to not let obvious telltale signs expose them, on a feature they created and made public.
Finally, the programs chosen for the study, “Old Boy” and “Hot Blood Dance Crew” were aired in 2018. The number of viewers for these out-of-favor shows is likely low and subject to inconclusive patterns given a small base.
Are iQIYI’s subscription numbers exaggerated?
Wolfpack alleged that iQIYI “must be misrepresenting its number of paying subscribers, the average membership period, or both.” The researchers provided an example of a theoretical membership-based growth company where the deferred revenue climb in line with the realized revenue.
Source: Wolfpack Research
Now, here’s where misdirection works its magic. It seems Wolfpack was trying to get readers to believe iQIYI’s deferred revenue came entirely from its membership income. However, iQIYI also earns from the sale of virtual currencies and ticketing, besides subscriptions.
In iQIYI’s 2019 Annual Report, the company explains its treatment of the revenue from the sale of consumable virtual gifts:
Revenue from the sale of consumable virtual gifts is recognized when consumed by the user, or, in the case of time-based virtual items, recognized ratably over the period each virtual item is made available to the user. Virtual currency sold but not yet consumed by the purchasers is recorded as “Customer advances and deferred revenue”.
The same annual report also states that deferred revenue includes income “in relation to content distribution, licenses of intellectual property and traffic support services to be provided to an equity investee.”
Source: Wolfpack Research
iQIYI’s barter sublicensing revenues are inflated?
Wolfpack Research charged that iQIYI reported artificially high barter sublicensing revenues by inflating the worth of its shows that were bartered. The researchers relied on the figures provided by “a former IQ employee who worked in content acquisition” that non-exclusive licenses “are typically worth RMB 1,000 to 5,000 per episode, or a maximum of up to RMB 20,000 for an extremely popular show.”
We have no idea when the former employee last worked at iQIYI (if he was even real) or in content acquisition. However, the content acquisition cost is ballooning at a frightening pace. In one article discussing the topic last year, the writer stated that due to the proliferation of online video platforms and the ensuing stiff competition, the cost of IP rights has exploded as much as 30 times.
RMB1,000 to 5,000 per episode just doesn’t pass the smell test. As early as in 2016, there was a hoo-ha in the entertainment sphere that a particular drama was selling for RMB8 million per episode. Sure, China’s top-earning actress Fan Bing Bing was the lead for the show but even mediocre dramas on a non-exclusive license shouldn’t be as low as purported by Wolfpack Research.
A hasty generalization to cast aspersions on iQIYI
Wolfpack Research sought to aspersions on iQIYI with regards to an acquisition it made:
“Before its acquisition by IQ, Skymoons had twice tried and twice failed to be acquired by two different Chinese listed companies (Jinya Tech, 300028.SZ and Ningbo Fubang, 600768.SHA); both of whom were subsequently charged with illegal or fraudulent market practices.”
Such acts belong to a type of fallacy called hasty generalization. Just because Skymoons was previously pursued by two companies that found themselves in trouble subsequently does not mean another acquirer will be involved in fraud. Furthermore, it was not uncommon in China for listed companies to acquire games businesses around the time of the purported deals.
Wolfpack discovered that Skymoons registered no new publication copyrights since the July 2018 acquisition. Instead, the researchers found only the distribution rights for four games under its name, all of which were published by other game companies. The report implied that since iQIYI CEO Gong Yu “only credits Skymoons for launching and adapting” the game titled The Croods, not designing it, it was a tacit acknowledgment of the limited capability of the acquired game company.
Wolfpack’s assertion sounds valid, given that iQIYI paid more than RMB2 billion for Skymoons. However, it is not a concrete proof that iQIYI overpaid. My contacts told me that in China, the ability to successfully launch and market games is highly valued.
Inflated advertising revenue by iQIYI?
Wolfpack Research alleged that iQIYI’s reported numbers in its SEC filings do not match what the company reported to the Shanghai branch of China’s State Administration of Industry and Commerce (‘SAIC’). Such mismatches have often been cited as “smoking gun evidence” for companies committing frauds.
Source: Wolfpack Research
However, discrepancies between the two sets of filings could be due to legitimate reasons. Thornhill Capital, a business services provider, listed some common explanations on its website:
- SAIC filings for a Chinese company with overseas operations will only indicate the Chinese portion of that business and not take into account the company’s global consolidated revenue and earnings.
- Each subsidiary of a Chinese company files its own report with SAIC, so it takes time and effort to manually consolidate these reports to reconcile to SEC filings.
- SAIC filings do not reflect cash kept in U.S. bank accounts to pay for expenses outside of China.
- There are fundamental differences between US Generally Accepted Accounting Principles (US GAAP) and those employed in China. For example, revenue recognition, derivative financial instrument accounting, impairment rules, and many other aspects of accounting are much more complicated under US GAAP, resulting in different financial results.
Advertisement of Kinder Joy on iQIYI’s platform prior to the start of the content.
Source: Screenshot by ALT Perspective
In the case of iQIYI, the company also has subsidiaries in Beijing (see the chart as follows). It is not known what is the contribution from its ex-Shanghai operations but it could explain the purported gap.
Source: iQIYI’s IPO prospectus
Misleading financial reporting by iQIYI?
Wolfpack concluded the report with a less than solid bombshell section titled: “Misleading Financial Reporting Creates the Appearance of a Cash Generative Company”. I have not come across investors who were under the illusion that iQIYI is a cash generator. It’s pretty obvious iQIYI is like its U.S. counterpart Netflix (NFLX), requiring loads of investments in content creation and licensing to attract and retain users.
The researchers argued that Netflix accounts for “all additions to steaming content assets” as cash outflows from operating activities, while iQIYI assigned the acquisition costs of the licensed copyrighted material as cash outflows from investing activities. The latter’s accounting treatment meant that the initial purchases have no negative impact on its operating cash flow (‘OCF’). Yet, the subsequent amortization has a positive impact on its OCF, giving investors the impression that iQIYI is a cash-generating company.
Again, I found this assertion reasonable given the possibility of misleading some investors. However, the cash flow statement is not something appearing in the fine prints. Most importantly, iQIYI has already stated in its 2019 Annual Report that the company “will adopt ASU 2019-02 on January 1, 2020, which the FASB issued in March 2019, and report cash flows for the decrease or increase of acquisition of licensed copyrights as ‘operating activities’ in the statement of cash flows, beginning with the period of adoption.” That means this will no longer be an issue from the next quarterly report.
Finally, I would like to share that the Chinese I chatted with on this topic told me they regarded Dan David, the founder of Wolfpack Research, as someone good at manipulating public sentiment. They felt he also lacked integrity. In the oft-mentioned documentary The China Hustle, where Dan played a leading role in exposing deceptive businesses, he concealed his identity and made loads of secret recordings at a few factories in rural China. Riding on his ‘discoveries’, he went on to warn about conniving Chinese executives in practically the whole of China. That’s too sweeping and despicable in their opinion.
iQIYI’s lack of a comprehensive rebuttal should not be taken as an admission of guilt. The company management either doesn’t see a need to waste their precious time on the (amateur) report or just like Qutoutiao, it wants to craft a convincing response and that takes time. The former is likely the case. In local chatters, CEO Yu Gong was said to have lamented: “邪不胜正，看最后谁赢” which means the righteous will prevail over the evil, let’s see who emerges triumphant in the end. Majority shareholders Baidu (BIDU), Xiaomi (XI), and Hillhouse Capital may even want it to stay silent so that they can increase their ownership in it at a lower price.
Dan asserted that the fraudulent practices were already present in its IPO prospectus in 2018. Given that the company has engaged in ballooning its actual numbers for some time, he reckoned that iQIYI is forced to maintain the false pretense as it needed to placate shareholders amid sustained heavy losses. If this is true, I found it strange that no short-seller before him have found fault with this supposedly ticking time-bomb.
Nevertheless, while many of the allegations by Wolfpack turned out to be untenable, it does not mean iQIYI is clean. Investors would need to do the proverbial homework and investigate further. In the meantime, iQIYI has rebounded quickly from the initial shock.
Disclosure: I am/we are long BIDU. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Editor’s Note: This article covers one or more microcap stocks. Please be aware of the risks associated with these stocks.